The Third PLeural Effusion And Symptom Evaluation (PLEASE-3) study: bendopnoea in patients with pleural effusion

Graphical abstract

Bendopnoea is common in pleural effusion, associated with larger effusions, worse breathlessness and shorter 6-min walk distance (6MWD), and predicts a greater improvement in 6MWD post-drainage. Some of the components of the illustrations were generated using Gemini 3 Flash Image (Google LLC, 2026), a multimodal generative AI model. The raw outputs were subsequently modified and re-composed by the authors to ensure clinical accuracy.

Abstract

Background

Bendopnoea (breathlessness on bending forward) is reported often by patients with pleural effusion, but its prevalence and clinical relevance are unexplored.

Methods

This study prospectively enrolled unselected patients with pleural effusion. Bendopnoea was assessed by time to breathlessness when bending patients forward for up to 60 s (60-s bend test (60s-BT)). Post-hoc analyses using a 30-s cut-off were also conducted.

Results

200 patients (median age 72 years; 61% male) were included; n=88 underwent therapeutic drainage (median (interquartile range (IQR)) 1.50 (1.11–2.47) L). Effusions were commonly malignant (42%) and occupied >25% of the hemithorax (74%). Most patients had bendopnoea on the 60s-BT (56%) and reported it “impacted daily activities” (74%). The estimated median time to bendopnoea was 43 (95% CI 30–59) s. Bendopnoea was more common with larger pleural effusions on chest radiographs (p=0.0332 using 60-s cut-off and p=0.0217 using 30-s cut-off). Patients with bendopnoea on the 60s-BT were more breathless (median (IQR) 48 (22–67) versus 30 (5–54) mm on a 100-mm Visual Analogue Scale; p=0.0002) and had shorter 6-min walk distance (6MWD) (median (IQR) 200 (128–333) versus 310 (223–371) m; p=0.0013). Similar relevance was found using a 30-s cut-off. The proportion of patients with bendopnoea on the 60s-BT significantly decreased from 62% to 32% post-drainage (p<0.001). Those with bendopnoea (versus those without) had significantly greater improvement in 6MWD post-drainage (median (IQR) 66 (2–101) versus 16 (−40–65) m; p=0.0165). Larger effusion (OR 5.08, 95% CI 1.25–20.66), inverted/flattened hemidiaphragm on ultrasonography (OR 3.93, 95% CI 1.32–11.68) and higher body mass index (OR 1.18, 95% CI 1.08–1.28) were associated with bendopnoea (60s-BT) in multivariable analysis.

Conclusion

Bendopnoea is a simple and worthwhile symptom to recognise in patients with pleural effusion.

Shareable abstract

This study found bendopnoea (breathlessness on bending forward) is common with pleural effusion and associated with larger effusion, worse dyspnoea and functional status. Those with bendopnoea had more significant benefits in walk distance post-drainage. https://bit.ly/4cJHczu

Introduction

Pleural effusion affects 24 million people worldwide each year and is characterised by abnormal accumulation of fluid within the pleural cavity. Breathlessness, particularly on exertion, is the most common symptom described in standard texts but is non-specific to pleural effusion. Patients often report a range of other symptoms, although many overlap their underlying comorbidities.

The PLeural Effusion And Symptom Evaluation (PLEASE) series aim to investigate the mechanism of breathlessness in pleural effusion. Results from PLEASE-1 and -2 showed that pleural effusion significantly alters the respiratory mechanics of both the ipsilateral and contralateral hemidiaphragms; changes that are reversible upon fluid drainage [1, 2]. Nonetheless, 27% of patients have no symptomatic benefits in breathlessness after fluid evacuation [1].

Bendopnoea (breathlessness on bending forward) is anecdotally reported by many patients with a pleural effusion, typically when putting on shoes and/or tying shoelaces. Bendopnoea has never been investigated in the context of pleural effusion. Its prevalence and relationship with pleural effusion have not been established. Importantly, the clinical usefulness and implications of recognising bendopnoea as an additional symptom of pleural effusion is unexplored.

The PLEASE-3 study aimed to investigate, for the first time, 1) the prevalence of bendopnoea in patients with pleural effusion, 2) the relationship of bendopnoea with characteristics of the pleural effusion and whether it can be improved with fluid drainage, and 3) if the presence of bendopnoea offers useful information to aid clinical management.

Methods

The protocol of the PLEASE-3 study has been published [3] and is summarised here.

This prospective study recruited 200 patients referred for assessment of a pleural effusion to the Respiratory Services of Sir Charles Gairdner Hospital, Perth, Australia and Westmead Hospital, Sydney, Australia between August 2022 and June 2024. The study was approved by the Sir Charles Gairdner and Osborne Park Health Care Group Human Research Ethics Committee (RGS0000004786) and is registered at the Australian New Zealand Clinical Trials Registry (ACTRN12622000465752). All patients signed an informed consent.

Participants

All patients with a pleural effusion identified by clinicians on chest radiographs or bedside ultrasound were triaged as potential study participants. Exclusion criteria included age <18 years, mechanical limitations in bending forward (e.g. spinal conditions or ascites), pregnancy/lactation and inability to consent or comply with the protocol.

Study measurements

Baseline demographics, comorbidities, body mass index (BMI; kg·m⁻²), smoking history and Eastern Cooperative Oncology Group performance status were recorded.

Pleural effusion

Pleural effusion characteristics including aetiology, size, fluid biochemistry and, where applicable, pleural drainage procedure were recorded.

The size of the pleural effusion on chest radiographs was graded from 0 to 5 by two assessors (grade 0=no fluid, grade 1=blunting of costophrenic angle, grade 2=fluid occupied <25% of the hemithorax, grade 3=fluid occupied 25–50% of the hemithorax, grade 4=fluid occupied 50–75% of the hemithorax and grade 5=fluid occupied >75% of the hemithorax) as per Light's categorisation of pleural effusion size on chest radiographs [4] and as used in the PLEASE-1 and -2 studies [1, 2]. Effusion sizes were categorised into three groups for all analyses: small effusions <25% of the hemithorax (grade 0–2), moderate effusions 25–50% of the hemithorax (grade 3) and large effusions >50% of the hemithorax (grade 4–5). The average grades of the assessors were used. The two assessors discussed any cases where their grades differed by more than one grade and the consensus grade was used.

Bendopnoea

This study employed a 60-s bend test (60s-BT) to provoke bendopnoea [5, 6]. Sitting in a chair, patients were instructed to bend forward at the waist and aim to touch their ankles. This position was maintained for 60 s or until the patient reported onset of bendopnoea. If bendopnoea developed within 60 s, the test was deemed positive, and the severity was determined as the time to development of bendopnoea. Those who did not develop breathlessness within 60 s were recorded as negative for bendopnoea, and time to bendopnoea development was censored as 60 s. The 60s-BT results were used for analyses of bendopnoea (i.e. positive/negative and time to bendopnoea development as severity). Post-hoc analyses using 30 s as the cut-off for bendopnoea were also presented.

A patient-reported bendopnoea questionnaire modified from Baeza-Trinidad et al. [6] was employed to determine the impact of bendopnoea on daily living [3].

Breathlessness

Breathlessness was measured using a validated 100-mm horizontal Visual Analogue Scale (VAS) [3, 7] previously used in malignant pleural effusion studies [8–10]. A minimal clinically important difference (MCID) of 14 mm was used [7].

Spirometry

Spirometry was completed in accordance with the joint American Thoracic Society/European Respiratory Society recommendations [11], and volumes of forced expiratory volume in 1 s (FEV₁) and forced vital capacity (FVC) were recorded.

Functional capacity

Functional capacity was measured using a standard 6-min walk test (6MWT) following international guidelines [12–14]. Patients who required oxygen in the baseline 6MWT were provided with supplemental oxygen at the same flow rate in the post-drainage test. The MCID was defined as 25 m [15].

Diaphragm characteristics

The shape and movement of the hemidiaphragm were assessed with patients sitting in an upright position using bedside ultrasonography as per published methods [16–18]. In accordance with the PLEASE-1 and -2 studies, the shape of the hemidiaphragm was classified as “normal” (dome-shaped), “flattened” or “inverted” and hemidiaphragmatic movement was classified as “normal”, “reduced”, “paradoxical” or “absent” [1, 2].

Post-drainage assessments

Patients who received therapeutic drainage (defined as >750 mL of pleural fluid removed) were included in the post-drainage analysis. Post-drainage assessments were performed between 24 and 72 h post-thoracentesis or when maximal drainage was reached in patients with chest drain insertion [1, 19].

Statistical methods

Data are summarised as median (interquartile range (IQR)) or as count (percentage). Kaplan–Meier analyses were used to analyse time to bendopnoea at baseline on the 60s-BT overall and the log-rank test was used to examine the impact of effusion size. Median times to bendopnoea are presented with 95% bootstrap confidence intervals. Mann–Whitney tests were used to compare median outcome measure scores (including breathlessness (VAS) and 6-min walk distance (6MWD)) between patients with and without bendopnoea on the 60s-BT at baseline and post-drainage. Wilcoxon signed-rank tests were used to assess within-group changes of VAS and 6MWD from pre- to post-drainage. McNemar's test was used to compare proportions with bendopnoea on the 60s-BT before and after drainage. Multivariable logistic regression was used to explore associations between baseline characteristics and the presence of bendopnoea on the 60s-BT with odds ratios and 95% confidence intervals reported. Analyses were repeated using 30 s as the bendopnoea cut-off time. Spearman correlation (r_s) coefficients were calculated to assess associations between severity of bendopnoea at baseline (i.e. time to bendopnoea when bending forward) and changes in breathlessness (VAS) and 6MWD scores post-drainage. All data were analysed using R statistical software [20].

Results

200 patients (median (IQR) age 72 (59–80) years; 61% male) were enrolled between August 2022 and June 2024 (table 1 and figure 1). Malignancy (42%) and pleural infections (17%) accounted for most of the pleural effusions. The pleural effusions were more commonly right-sided (48%) and occupied 25–50% of the hemithorax on chest radiographs, although the cohort covered all sizes from blunting of the costophrenic angle to complete “white-out” (table 1).

TABLE 1.

Baseline characteristics (n=200)

Age (years)	72 (59–80)
Male	122 (61)
BMI (kg·m−2) (n=190)	26 (23–30)
Smoking status (n=197)
Never	85 (43)
Current	23 (12)
Former	89 (45)
Pack-years (n=99)	20 (8–40)
ECOG performance status
0	60 (30)
1	71 (36)
2	34 (17)
3	33 (16)
4	2 (1)
Comorbidities
Malignancy	121 (61)
Cardiac	107 (54)
Respiratory	56 (28)
Diabetes	27 (14)
Renal failure	23 (12)
Liver disease	10 (5)
Pleural effusion
Side of effusion
Right	95 (48)
Left	72 (36)
Bilateral	33 (17)
Size of effusion (n=192)#
<25% of hemithorax	50 (26)
25–50% of hemithorax	81 (42)
>50% of hemithorax	61 (32)
Cause of effusion
Malignant	84 (42)
Pleural infection	33 (17)
Transudative	28 (14)
Unknown	25 (12)
Other	30 (15)
Baseline physiology
Breathlessness VAS score (mm) (n=199)	40 (15–63)
6MWD (m) (n=157)	248 (160–354)
FVC (% pred) (n=185)	43 (32–58)
FEV1 (% pred) (n=185)	43 (26–56)

Data are presented as median (interquartile range) or n (%). BMI: body mass index; ECOG: Eastern Cooperative Oncology Group; VAS: Visual Analogue Scale; 6MWD: 6-min walk distance; FVC: forced vital capacity; FEV₁: forced expiratory volume in 1 s. ^#: according to Light's categorisation of pleural effusion size on chest radiographs [4].

FIGURE 1.

Participation flowchart. ^#: participants were recruited from two sites: Sir Charles Gairdner Hospital, Perth, Australia (n=165) and Westmead Hospital, Sydney, Australia (n=35); ^¶: participants were assessed as inpatients (n=116) and outpatients (n=84); ⁺: 12 participants did not have post-drainage assessments (participant was under palliation or died (n=6), required urgent surgery (n=4) or declined further assessments (n=2)).

Bendopnoea: prevalence

Bendopnoea developed in 111 patients (56%) within 60 s when asked to bend forward in the 60s-BT (“positive bendopnoea”). The estimated median time to developing bendopnoea was 43 (95% CI 30–59) s in the entire cohort but half of those with bendopnoea (n=56) tolerated <16 s. 86 patients (43%) developed bendopnoea within 30 s.

Impact of bendopnoea on daily living

In the bendopnoea questionnaire [3], 74% (n=148) of patients reported that bendopnoea “impacted their daily lives”, 19% (n=37) reported “no” and 8% (n=15) reported “don't know”. Of those who reported having bendopnoea, 16% (n=23) indicated it had a “severe” impact on their daily life and 44% (n=65) indicated a “moderate” impact.

Bendopnoea: clinical relevance

Patients with bendopnoea on the 60s-BT (n=111) were more breathless compared to those without bendopnoea (n=88): median (IQR) breathlessness VAS score 48 (22–67) versus 30 (5–54) mm, respectively (p=0.0002) (figure 2). They also walked significantly shorter 6MWD than those without bendopnoea: median (IQR) 200 (128–333) versus 310 (223–371) m (p=0.0013) (figure 3).

FIGURE 2.

Baseline breathlessness Visual Analogue Scale (VAS) score in patients with bendopnoea on the 60-s bend test (n=111) versus without bendopnoea (n=88). Those with bendopnoea were significantly more breathless. Median and interquartile range are indicated.

FIGURE 3.

Baseline 6-min walk distance (6MWD) was significantly lower in those with bendopnoea on the 60-s bend test (n=84) than those without bendopnoea (n=73). Median and interquartile range are indicated.

Positive bendopnoea on the 60s-BT was associated with increasing sizes of pleural effusions: 62% were positive bendopnoea in those with effusion sizes >50% of the hemithorax compared to 58% in those with effusion sizes 25–50% of the hemithorax and 23% in those with effusion sizes <25% of the hemithorax (p=0.0332) (table 2). Median time to development of bendopnoea exhibited a size-dependent relationship with the pleural effusion on radiographs: 30.33 (95% CI 25.47–59) s versus 38.97 (95% CI 23–46) s versus >60 s (77% were censored at 60 s) for effusion sizes >50% of the hemithorax, 25–50% of the hemithorax and <25% of the hemithorax, respectively (p=0.0546). Positive bendopnoea was also associated with large sizes of pleural effusions in the post-hoc analyses using an alternative 30-s cut-off (p=0.0217) (table 2).

TABLE 2.

Baseline positive bendopnoea versus effusion size: bendopnoea is more common with increased size of pleural effusion on chest radiographs

Size of effusion# (n=109)	Bendopnoea with 60-s cut-off (n=111)	p-value comparing proportions of those with bendopnoea	Bendopnoea with 30-s cut-off (n=86)	p-value comparing proportions of those with bendopnoea
<25% of hemithorax	3 (23)		1 (8)
25–50% of hemithorax	68 (58)	0.0332¶	53 (45)	0.0217¶
>50% of hemithorax	38 (62)		30 (49)

Data are presented as n (%), unless otherwise stated. ^#: according to Light's categorisation of pleural effusion size on chest radiographs [4]; ^¶: Chi-squared (p-value represents comparison of all three effusion sizes for each cut-off).

Bendopnoea: associations of clinical responses to drainage of pleural effusion

Therapeutic drainage (>750 mL) was performed in 88 patients who were clinically indicated as per standard care; 12 were unable to perform post-drainage assessments (see caption to figure 1 for reasons). In the 76 patients who completed post-assessments, a median (IQR) volume of 1.50 (1.11–2.47) L was removed via an intercostal catheter (38%), thoracentesis (33%) or indwelling pleural catheter (29%). Breathlessness significantly improved after therapeutic fluid drainage, as evident by a median (IQR) decrease in breathlessness VAS scores of 24 (5–42) mm (p<0.001). The median (IQR) 6MWD also improved by 46 (0–82) m (p=0.0002) post-drainage in the 53 (75%) patients who were able to complete a 6MWT.

For those that had measures both pre- and post-drainage, the proportion with bendopnoea on the 60s-BT significantly decreased from 62% to 32% (p<0.001). Improvement in 6MWD was significantly greater in those with baseline bendopnoea on the 60s-BT (n=30) compared to those without (n=23): median (IQR) improvement 66 (2–101) versus 16 (−40–65) m, respectively (p=0.0165) (table 3). These results were consistent with post-hoc analyses using the 30-s cut-off (p=0.0019) (table 4). A significant correlation (r_s= −0.46, p=0.0006) was shown between worse baseline bendopnoea (on the 60s-BT) and greater improvement in 6MWD post-drainage.

TABLE 3.

Bendopnoea with 60-s cut-off versus 6-min walk distance (6MWD) change and breathlessness Visual Analogue Scale (VAS) score change post-drainage in patients who had >750 mL of pleural fluid removed

	Baseline bendopnoea with 60-s cut-off		p-value#
	No bendopnoea	Bendopnoea
6MWD baseline (m) (n=67)	285 (218–365) (n=31)	163 (103–308) (n=36)	0.0157
6MWD change (m) (n=53)	16 (−40–65) (n=23)	66 (2–101) (n=30)	0.0165
Breathlessness VAS score baseline (mm) (n=88)	34 (15–62) (n=34)	51 (34–69) (n=54)	0.0075
Breathlessness VAS score change (mm) (n=76)	−22 (−32– −5) (n=29)	−27 (−49– −4) (n=47)	0.4413

^#: Mann–Whitney.

TABLE 4.

Bendopnoea with 30-s cut-off versus 6-min walk distance (6MWD) change and breathlessness Visual Analogue Scale (VAS) score change post-drainage in patients who had >750 mL of pleural fluid removed

	Baseline bendopnoea with 30-s cut-off		p-value#
	No bendopnoea	Bendopnoea
6MWD baseline (m) (n=67)	288 (218–378) (n=38)	160 (100–202) (n=29)	0.0003
6MWD change (m) (n=53)	16 (−40–65) (n=29)	78 (10–115) (n=24)	0.0019
Breathlessness VAS score baseline (mm) (n=88)	39 (15–62) (n=43)	52 (34–74) (n=45)	0.0105
Breathlessness VAS score change (mm) (n=76)	−18 (−32– −2) (n=38)	−33 (−50– −9) (n=38)	0.0954

^#: Mann–Whitney.

Those with bendopnoea had improvement in breathlessness VAS scores post-drainage: median (IQR) decrease −27 (−49– −4) mm (p=0.4413). Although patients with bendopnoea had worse baseline breathlessness VAS scores than those without, there were no significant intergroup differences in the magnitude of improvement (p=0.4413) (table 3). These results were consistent with post-hoc analyses using the 30-s cut-off (p=0.0954) (table 4). There was an inverse correlation between time to bendopnoea at baseline (60s-BT) and breathlessness VAS score post-drainage (r_s= −0.33, p=0.0042).

Bendopnoea: associations with patient baseline characteristics

Larger pleural effusion (OR 5.08, 95% CI 1.25–20.66), abnormal shape (i.e. inverted/flattened) of the ipsilateral hemidiaphragm (OR 3.93, 95% CI 1.32–11.68) and higher BMI (OR 1.18, 95% CI 1.08–1.28, per 1-unit increase) were significantly associated with bendopnoea at baseline on the 60s-BT in multivariable analyses of the entire cohort (table 5). Results were similar using the 30-s cut-off (table 5).

TABLE 5.

Associations with baseline bendopnoea using multivariable logistic regression

Effect	60-s bendopnoea cut-off	30-s bendopnoea cut-off
Age (continuous)#	1.01 (0.97–1.04)	0.99 (0.96–1.03)
Sex
Female versus male	0.71 (0.30–1.70)	0.85 (0.32–2.23)
BMI (continuous)#	1.18 (1.08–1.28)	1.26 (1.14–1.40)
Effusion type
Transudative versus exudative	0.62 (0.17–2.25)	1.33 (0.32–5.54)
Unknown versus exudative	1.40 (0.27–7.31)	0.97 (0.14–6.61)
Diaphragm shape (baseline)
Normal/domed versus inverted/flattened	3.93 (1.32–11.68)	10.29 (2.65–39.96)
Diaphragm movement (baseline)
Not moving/paradoxical versus normal/reduced	2.23 (0.76–6.54)	4.47 (1.25–15.98)
Effusion size (baseline)¶
25–50% versus <25% of hemithorax	3.56 (1.02–12.47)	1.96 (0.49–7.76)
>50% versus <25% of hemithorax	5.08 (1.25–20.66)	4.00 (0.80–19.95)
Baseline breathlessness VAS score (mm)+	1.57 (1.00–2.47)	1.81 (1.08–3.03)
Baseline 6MWD (m)+	0.72 (0.43–1.21)	0.45 (0.25–0.84)
Baseline FEV1 (% pred)+	0.68 (0.13–3.48)	0.33 (0.06–1.96)
Baseline FVC (% pred)+	1.08 (0.20–5.79)	1.57 (0.27–9.27)

Data are presented as odds ratios with 95% confidence intervals. ^#: odds ratio for 1-unit increase in continuous variable; ^¶: according to Light's categorisation of pleural effusion sizes on chest radiographs [4]; ⁺: odds ratio for 1-sd increase in continuous variable. BMI: body mass index; VAS: Visual Analogue Scale; 6MWD: 6-min walk distance; FVC: forced vital capacity; FEV₁: forced expiratory volume in 1 s.

Discussion

This study is the first to present bendopnoea as a symptom in patients with pleural effusion and highlight its clinical relevance. Our data demonstrate that bendopnoea is common in this cohort, provoked in more than half of all patients and reported by 74% as impacting daily activities. Clinician awareness is important for explaining why patients may find simple tasks difficult (e.g. putting on shoes or tying shoelaces) and plan appropriate supportive care (e.g. occupational therapy for task modification). Our findings are hypothesis generating and support a potential causal relationship of bendopnoea with pleural effusion; there was significant improvement of bendopnoea in patients after fluid drainage as well as a size-dependent relationship of the effusion with bendopnoea. Importantly, bendopnoea was associated with significant functional impairment on the 6MWT and better improvement post-drainage.

Clinicians need to recognise bendopnoea as an important symptom of pleural effusion. Although bendopnoea has not previously been associated with pleural diseases, our data found a very high prevalence, and those affected often graded bendopnoea as having a “severe” or “moderate” impact on their daily activities.

There are no standardised tests for bendopnoea. Using a quick and simple “bend test” for up to 60 s, we were able to elicit clinically relevant information. Our 60-s cut-off was arbitrary but represents the approximate time to put on shoes and tie shoelaces (activities most often reported by patients to cause bendopnoea). It is possible a longer duration (>60 s) of bending may provoke bendopnoea in even more patients, as would asking patients to perform activities replicating real-life movements while bending forward. The first characterisation of bendopnoea [21] and some other published studies of the symptom employed a 30-s cut-off [5, 6, 22]. In our study, results from either a 60- or 30-s cut-off were very similar.

Our findings provide support for a potential causal relationship of bendopnoea with pleural effusion, as evident by 1) the severity of bendopnoea is strongly related to the radiographic size of the pleural effusion and 2) the significant improvement of bendopnoea upon therapeutic fluid removal. However, bendopnoea is not unique to pleural effusion, and has been described in other cardiorespiratory diseases, such as heart failure and pulmonary hypertension [21–25]. It is possible that other comorbidities also add to the severity of bendopnoea in some patients with pleural effusion. In our unselected cohort, 54% of patients had cardiac comorbidities and 14% of effusions were transudative; however, multivariable analyses showed no association between effusion type (transudative versus exudative) and the presence of bendopnoea at baseline.

Nonetheless, the presence of bendopnoea is associated with a larger effusion, abnormal hemidiaphragmatic appearances and significant breathlessness as well as functional restriction. In our prior PLEASE-1 study, 27% of patients had no significant symptom benefits despite therapeutic fluid drainage (median 1.7 L) [1]. In the current PLEASE-3 study, patients with bendopnoea had significantly better improvement in 6MWD when fluid was drained. The usefulness of bendopnoea should be incorporated into the decision whether to perform therapeutic drainage of pleural effusions for symptomatic relief.

The underlying pathophysiology of bendopnoea remains unclear, although various mechanisms have been postulated in the literature [21, 24]. It is possible that akin to breathlessness, bendopnoea in various conditions may be multifactorial and may have different pathophysiology. The interplay of heart–lung mechanics, including increased venous return leading to pulmonary congestion, increased left ventricular filling pressures and altered diaphragmatic mechanics during bending, may be possible contributors to bendopnoea. In our cohort, the impact of pleural fluid weight on the diaphragm may also contribute to the experience of bendopnoea, as demonstrated by the improvement in bendopnoea after fluid evacuation. The implications of gravity on the hemidiaphragm mechanics, strength of respiratory muscles and perfusion of upper lobes of the lungs when bending forward are yet to be explored in the context of effusion-related bendopnoea.

In the multivariable analysis, correlates of bendopnoea included the size of effusion, hemidiaphragm movement and BMI. This contrasts with previous studies showing effusion size was not reliably related to breathlessness [26]. The prior PLEASE-1 and -2 studies demonstrated that pleural effusion affects both ipsilateral and contralateral hemidiaphragms in their shape and movement [1, 2]. Flattening or inversion of the ipsilateral hemidiaphragm is an independent predictor for breathlessness relief post-drainage [1, 2]. BMI has been inconsistently associated with bendopnoea [5, 6, 24, 25, 27, 28], and in our study it was weakly associated with presence of bendopnoea.

This study has limitations. First, bendopnoea on the 60s-BT relied on self-reporting breathlessness upon bending, which can be influenced by the Hawthorne effect. Nonetheless, correlation of the time to bendopnoea with objective parameters (e.g. size of effusion) suggests that it is a reliable method. Second, discomfort or pain associated with underlying pleural diseases may have confounded the 60s-BT. Third, the optimal time to assess for bendopnoea post-drainage is unknown. Prior studies have shown that symptomatic benefit of pleural drainage may peak at different times post-drainage. We chose the timing of 24–72 h as previous studies have shown physiological outcomes in patients could worsen in the first few hours post-drainage and improvements are more consistent by 24 h [1, 19].

Bendopnoea, and its impact on daily activities, is currently unrecognised in patients with pleural effusion. Clinicians should be aware of the clinical implications when patients report bendopnoea, especially the likelihood of functional benefits of fluid drainage. Bendopnoea in pleural effusion is a worthwhile topic for future research. Optimising/standardising the diagnostic/testing criteria in pleural effusion patients is important. Higher breathlessness VAS score is associated with prognosis in malignant effusion patients; whether bendopnoea has survival implications should be explored. Establishing the pathophysiology of bendopnoea may enhance our understanding of breathlessness in pleural diseases.