AJTCCM VOL. 29 NO. 4 2023 171
ORIGINAL RESEARCH: BRIEF REPORTS
Tuberculous pleural eusion (TPE) and malignant pleural eusion
(MPE) constitute the majority of exudative pleural effusions in
Malaysia.[1] Histopathological analysis is diagnostic for these
conditions,[2] while pleural uid cytology has a sensitivity of only
58.2% for MPE.[3] Clumps of tissue from chest drains are of diagnostic
value in the setting of cytology-conrmed MPE,[4] but the diagnostic
role of such macroscopic tissue in cytology-negative eusions remains
to be elucidated. In the present study, tissue termed pleural uid
residue (PFR) was obtained through direct ltration. e objectives
of the study were to determine: (i) histopathological correlation of
PFR with thoracoscopic pleural biopsy in diagnosing TPE and MPE;
and (ii) the cellularity, architecture and cytomorphology of PFR,
and whether immunohistochemical (IHC) staining was feasible. We
discuss the ndings in 18 cases.
This single-centre, prospective observational cohort study was
conducted on consecutive adult inpatients with cytology-negative
exudative pleural eusion who underwent medical thoracoscopy in
our respiratory unit from 6 September 2022 to 14 December 2022.
We excluded patients with bacterial empyema and those with a pre-
existing chest drain. A histopathological diagnosis of TPE in our setting
is dened as histological examination identifying granulomatous
inammation, or direct visualisation of acid-fast bacilli along with a
typical patient history.
During medical thoracoscopy, an initial volume of 500 mL of
pleural uid was suctioned aer trocar insertion, and the drain bottle
was removed for immediate ltering. e suctioned pleural uid was
extracted from the drain bottle and ltered through sterile 16-ply,
19 × 15 mesh size cotton gauze, which yielded PFR. e PFR was
gently rolled into a single mass, placed on a lter paper to air-dry for 2
minutes, and nally placed into a lter-paper envelope. is was then
xed in formalin and transported to the histopathology laboratory for
further processing and analysis. e act of rolling the ltered PFR into
a clump pays homage to the tissue coagulum clot cell block method of
transbronchial needle aspiration.[5]
Fig.1 compares the histopathological appearance of pleural tissue
obtained through pleural biopsy (A and B) and from PFR (C and
Pleural uid residue as a diagnostic tool for cytology-negative
malignant pleural eusion: A proof-of-concept study
L E Nyanti, MRCPI ; N-C Huan, FRCP; H Y Ramarmurty, MRCP; T Renganathan, MPath; M A bin Abdul Aziz, MPath;
J L Low, MBBS; K T Rosli, MRCPI; S Letcheminan, MRCP; M G Lansing, MRCP; K K Sivaraman Kannan, FRCP
Medical Department, Faculty of Medicine and Health Sciences, Universiti Malaysia Sabah, Kota Kinabalu, Malaysia
larrynyan[email protected]du.my
Department of Respiratory Medicine, Queen Elizabeth Hospital, Kota Kinabalu, Malaysia
Pathology Department, Queen Elizabeth Hospital, Kota Kinabalu, Malaysia
Department of Respiratory Medicine, Queen Elizabeth Hospital, Kota Kinabalu, Malaysia
Medical Department, Faculty of Medicine and Health Sciences, Universiti Malaysia Sabah, Kota Kinabalu, Malaysia
Department of Respiratory Medicine, Queen Elizabeth Hospital, Kota Kinabalu, Malaysia
Pleural uid residue, or macroscopic tissue, circulating freely in the pleural uid obtained through direct ltration, may carry diagnostic
histopathological information. We aimed to determine the histopathological concordance of pleural uid residue in diagnosing TPE and
MPE, compared with conventional pleural biopsy. is was a prospective cohort study of consecutive inpatients with cytology-negative
exudative eusion who underwent pleuroscopy and had their initial suctioned pleural uid ltered for residue samples. Pleural uid residue
demonstrated malignant cells in four out of seven cases of pleural biopsy-conrmed malignancy. Pleural uid residue has comparable
cytomorphology but reduced cellularity compared with pleural biopsy. No tuberculous histological features were present in the pleural uid
residue samples. In this preliminary study pleural uid residue provided histopathological information for malignant pleural eusion, but
no incremental diagnostic information for tuberculous eusion. However larger and more denitive studies are required to clarify these
ndings, and to explore the utility and suitability of pleural uid residue for mutational analysis
Keywords. Pleural uid residue, diagnostic accuracy, clump, malignant plural eusion.
Afr J Thoracic Crit Care Med 2023;29(4):e1149. https://doi.org/10.7196/AJTCCM.2023.v29i4.1149
Study synopsis
What this study adds. is study demonstrates the potential of pleural uid residue as a non-invasive diagnostic method for conrming
malignancy in cytology-negative exudative eusion.
Implications of the ndings. In resource-limited settings or patients contraindicated for pleural biopsy, pleural uid residue may provide
a viable diagnostic alternative; however, this observation needs further validation.
172 AJTCCM VOL. 29 NO. 4 2023
ORIGINAL RESEARCH: BRIEF REPORTS
D). Of the 18 cases, 17 yielded conrmatory pleural biopsy samples
(Supplementary Table 1, available online at https://www.samedical.org/
le/2143). ere was positive correlation in 4 out of 7 MPEs (57.1%).
Six patients were diagnosed with adenocarcinoma, of whom 4 (66.7%)
had PFR consistent with malignancy; one did not have IHC staining.
One patient had squamous cell carcinoma on biopsy, but PFR only
demonstrated inammatory cells. Histological examination of PFR
revealed malignant columnar cells trapped within brin or blood clot,
either arranged in a glandular pattern or singly distributed (Fig.1D). e
cytomorphology of PFR and its feasibility for IHC staining are similar
to those of pleural biopsy samples, although PFR has less cellularity
and poorer architectural visualisation (Fig.1A andC). We did not do
mutational analysis.
Four patients had TPE on pleural biopsy, but their PFR demonstrated
only inammatory cells. Diagnoses of uraemic pleuritis (n=1) and
parapneumonic eusion (n=5) were made based on clinical history
and examination, combined with absence of features of tuberculosis on
pleural biopsy. e overall negative correlation was 6 out of6 PE/TPE
cases. Both PFR and thoracoscopic pleural biopsy were concordant in
terms of the presence of inammatory cells, but PFR added no further
information.
In this proof-of-concept study, we demonstrated that PFR
contributes valuable histopathological information in the diagnosis
of cytology-negative malignant pleural effusion. Its fibrin-based
morphology is comparable to that of chest drain clumps previously
described.[4] PFR is morphologically comparable to pleural biopsy,
and immunohistochemical staining is feasible. Despite having less
cellularity, it is still theoretically sucient for mutational analysis.
In contrast to pleural fluid cytology, in which pleural fluid is
centrifuged to exclude any debris or sediments and subsequently
smeared for analysis, diagnostic information from PFR is obtained from
the debris found in pleural uid. PFR also diers from a cytology cell
block in that it does not require additives such as plasmaandthrombin
to enmesh the cellular material. Furthermore,PFR can be processed at
the patient’s bedside and immediately xed in formalin. is technique
could potentially negate the need for pleural biopsy in suspected
malignant pleural eusion, which would be particularly helpful in
patients who are not suciently stable formedical thoracoscopy, or
those who are far from a tertiary centre with the necessary equipment
to perform medical thoracoscopy.
It is likely that granulomas do not exist in TPE, as they are formed by
a lymphocyte-driven, macrophage-initiated immune reactionwhich
requires blood supply that is only present within the pleura.[6]
One would not expect to see acid-fast bacilli on histopathological
examination, given that tuberculous fluid is paucibacillary as a
resultof T-helper cell compartmentalisation and eective containment
of tuberculous bacilli.[2]
The present study is not without limitations. First, inconsistent
presence of PFR on ltration raises questions as to whether the ltration
method could be improved, or whether brin-containing residue is
uniformly formed in all kinds of malignant eusion; and if so, how
soon it forms. It is conceivable that a smaller mesh size could improve
PFR yield by capturing smaller residue, but this needs to be studied
further. Second, mutational analysis was not carried out. ird, the
potential role of PFR being ltered from newly inserted chest drains
was not explored in our study. Fourth, a small sample size and single-
centre setting reduces the generalisability of the ndings. Fih, the
usefulness of PFR for other types of malignancy is unproven, as only
adenocarcinoma was diagnosed in this study. ese limitations will be
addressed in a future pilot study.
PFR provides valuable diagnostic information to assist in the diagnosis
of malignant pleural eusion, but does not carry the histopathological
information necessary to diagnose pleural tuberculosis. Head-to-head
studies on the usefulness of PFR and cytology cell block may elucidate
PFR’s diagnostic potential. Future studies are needed to explore its role
in patients with newly inserted chest drains or non-adenocarcinoma
malignancy, and its suitability for mutational analysis.
Declarations. Written informed consent was obtained from the patients
for publication of this article and the accompanying images. A copy of
the written consent is available upon request from the Editor-in-Chief of
this journal. is study was approved by the Medical Research & Ethics
Committee of Malaysia (NMRR ID-22-02491-IED).
Acknowledgements. None.
Author contributions. Conceptualization: LEN, HYR, NCH.
Methodology: LEN, HYR, NCH. Formal analysis: TR, MABAA.
Investigation: LEN, NCH, HYR, JLL, KTR, SL, MGL. Data curation:
JLL, LEN, MGL, KTR, NCH, HYR. Writing – Original dra: LEN, NCH,
HYR. Writing-Review and Editing: HYR, NCH, KKSK, TR, MABAA, SL.
Supervision: KKSK, HYR, NCH
Funding. None.
Conict of interest. None.
1. Kannan SK, Lin WJ, Teck TS, Azizi ARJ. Pleuroscopy: Early experience in an East
Malaysian state with high tuberculosis prevalence. J Bronchology Interv Pulmonol
2009;16(4):250-253. https://doi.org/10.1097/LBR.0b013e3181ba730a
2. Shaw JA, Diacon AH, Koegelenberg CFN. Tuberculous pleural eusion. Respirology
2019;24(10):962-971. https://doi.org/10.1111/resp.13673.
Fig.1. Histopathological comparison between pleural tissue obtained
through pleural biopsy (A and B) and from PFR (C and D). Both samples
displayed strong and diuse immunoreactivity for TTF-1 (Aand C, 20×).
Under H&E staining, PFR demonstrated satisfactory cellularity, albeit
admixed with brin (D). Cytomorphological features were comparable and
characterised by malignant cells arranged in cords, nests and glandular
structures, with round to oval nuclei showing vesicular chromatin texture,
conspicuous nucleoli and a modest amount of eosinophilic cytoplasm
(B and D, H&E, 40×). (PFR = pleural uid residue; TTF-1 = thyroid
transcription factor-1; H&E = haematoxylin and eosin.)
AJTCCM VOL. 29 NO. 4 2023 173
ORIGINAL RESEARCH: BRIEF REPORTS
3. Kassirian S, Hinton SN, Cuninghame S, etal. Diagnostic sensitivity of pleural uid
cytology in malignant pleural eusions: Systematic review and meta-analysis. orax
2023;78(1):32-40. https://doi.org/10.1136/thoraxjnl-2021-217959
4. Louw A, Sidhu C, Fitzgerald DB, Creaney J, Chai SM, Lee YCG. Clump material within
drainage chest tubes contains diagnostic information: A proof-of-concept case series.
Eur Respir J 2021;57(3):2003248. https://doi.org/10.1183/13993003.03248-2020
5. Yung RCW, Otell S, Illei P, etal. Improvement of cellularity on cell block preparations
using the so-called tissue coagulum clot method during endobronchial ultrasound-
guided transbronchial ne-needle aspiration. Cancer Cytopathol 2012;120(3):185-
195. https://doi.org/10.1002/cncy.20199
6. Vorster MJ, Allwood BW, Diacon AH, Koegelenberg CFN. Tuberculous pleural
eusions: Advances and controversies. J orac Dis 2015;7(6):981-991. https://doi.
org/10.3978/j.issn.2072-1439.2015.02.18
Submitted 5 June 2023. Accepted 10 September 2023. Published x November 2023.
