Pulmonologist performed ultrasound guided fine needle aspiration of lung lesions

doi:10.21203/rs.3.rs-2193157/v1

Background and objective:

Lung cancer is increasingly common and accurate diagnosis is important for personalized treatment. Ultrasound guided percutaneous fine needle aspiration is a useful method to obtain specimen for histological diagnosis of peripheral lung lesions. The aim of this study is to evaluate the diagnostic accuracy and complication rate of the procedure performed by pulmonologists. The result is compared with ultrasound guided core needle biopsy performed by radiologists.

Methods:

We retrospectively evaluated the diagnostic accuracy and complication rate of pulmonologist-performed ultrasound guided FNA of lung lesions in the period of 1st August 2019 to 30th June 2021 (pulmonologist group) and radiologist-performed ultrasound guided CNB of lung lesions in the period of 1st January 2010 to 31st December 2014 (radiologist group). Logistic regression analysis was used to identify independent influence factors associated with diagnostic accuracy in pulmonologist group and in combining both groups.

Results:

In a 23-month period, pulmonologists in a tertiary center performed 113 episodes of ultrasound guided fine needle aspiration for peripheral lung lesions. The diagnostic accuracy and complication rate were 80.4% and 5.3% respectively, comparable to 86.8% and 7.4% in a historical cohort consisting of 68 episodes of ultrasound guided core needle biopsy performed by radiologists in the same hospital. Lung lesions of upper lobe location were predictive of successful diagnosis.

Conclusion:

Ultrasound guided fine needle aspiration by pulmonologist is an easily accessible and reliable method to obtain specimen for histological diagnosis.

pulmonologists performed

ultrasound guided

fine needle aspiration

peripheral lung lesion

Lung mass is a common finding in thoracic imaging(1). It is expected that pulmonologists will receive increasing number of referrals for lung mass, which poses challenge on resources for diagnostic procedures. The American College of Chest Physicians evidence-based clinical practice guidelines on diagnosis and management of lung cancer recommend that the diagnosis should be confirmed by the least invasive method(2). Typically, histological diagnosis was obtained by bronchoscopic methods for centrally located lung lesion(3) and by image guided percutaneous biopsy for peripherally located lung lesion(4). For lung lesion abutting the chest wall without aerated lung between the lesion and the pleura, ultrasound serves as a good image guidance(5). Ultrasound guided fine needle aspiration of lung lesions was proven to be safe in the hands of pulmonologists(6). It is a reliable method to obtain histology(7), even in patients with cancer complications such as superior vena cava syndrome(8). We retrospectively evaluated the diagnostic accuracy and complication rate of pulmonologist-performed ultrasound guided FNA of lung lesions in the period of 1^st August 2019 to 30^th June 2021 (pulmonologist group) and radiologist-performed ultrasound guided CNB of lung lesions in the period of 1^st January 2010 to 31^st December 2014 (radiologist group).

Objectives

Primary Objective:

To evaluate the diagnostic accuracy and complication rate of pulmonologist-performed ultrasound guided fine needle aspiration and radiologist-performed ultrasound guided core needle biopsy of peripheral lung lesions

Secondary Objectives:

To analyze factors that predict diagnostic accuracy by ultrasound guided FNA and CNB of lung lesions

a) Subjects

The full list of patients in the pulmonologist group was kept by the pulmonologists. The historical control of radiologist-performed ultrasound guided CNB of lung lesions in the period of 1^st January 2010 to 31^st December 2014 (radiologist group) was chosen for comparison of diagnostic accuracy and complication rate. The list of patients in the control group was obtained by searching the Clinical Data Analysis and Reporting System (CDARS) in Hospital Authority with the keyword of ‘fine needle aspiration biopsy of lung’ and ‘percutaneous fine needle aspiration biopsy of lung with imaging guidance’ (ICD code 33.26). Among the searched patients, cases performed by radiologists under ultrasound guidance would be recruited in the radiologist group as the historical control.

b) Diagnostic category

Cases of ultrasound guided sampling of lung lesions were classified as true positive, true negative, false positive and false negative. Cases with histological diagnoses of specific neoplastic conditions, infectious diseases or benign conditions which had consistent clinical course for 6 months after the procedure were defined as true positive. False negative was defined as necrotic material, quantity insufficiency, atypical cells or suspicious of malignancy in the FNA result, but the definitive diagnosis was subsequently made by other methods such as bronchoscopy, CT guided biopsy, surgical biopsy or microbiological culture. True negative was defined as non-diagnostic cytology with stable or receding lesions on follow up. Cases with clinical course inconsistent with tissue sampling suggested diagnoses were regarded as false positive. The diagnoses were categorized as primary lung cancers (with different histological subtypes), metastases, hematological malignancies, malignancy not otherwise specified, infection, benign lesions and non-diagnostic.

c) Study design

Diagnostic accuracy is defined as the proportion of true positive and true negative in all evaluated cases. The sensitivity, specificity, positive predictive value and negative predictive value would be calculated. Complications consist of pneumothorax and hemoptysis. Pneumothorax is defined as the presence of air in the ipsilateral pleural cavity on chest X ray within 24 hours after the procedure. Hemoptysis is defined as patient reported blood-stained sputum within 24 hours after FNA. The predictive factors that would be analyzed include age, gender, smoking status, lesion size and location.

d) Statistical analysis

The 95% confidence interval and p-value of diagnostic accuracy, sensitivity, specificity, positive predictive value and negative predictive value of both pulmonologist and radiologist group were calculated. Pearson’s chi-squared test and Fisher’s exact test were employed to test for difference between groups of categorical variables. Independent t-test and Mann-Whitney U test were used to test for difference between groups of continuous variables. Logistic regression analysis was used to identify independent influence factors associated with diagnostic accuracy in pulmonologist group and in combining both groups. Statistical analysis was conducted using SPSS version 27. A p-value <0.05 was considered as statistically significant.

e) The ultrasound guided fine needle aspiration

Patients were instructed to sit upright, lie supine or lie lateral. The ultrasound machine Venue 40® or Venue 50® (General Electric, Boston, USA) with curvilinear probe was used to locate the lesion. Color Doppler was employed to check for vessels running through the lesion. The operators would avoid puncturing the vessels during FNA. Under real time ultrasound guidance and aseptic technique, 5ml 2% lignocaine was injected percutaneously for local anesthesia. A 22-gauge (22G) spinal needle Spinocan® (B. Braun, Melsungen, Germany) was inserted and the needle tip position within the target lung lesion was confirmed by real time ultrasound exam in the short axis or long axis (figure 1). The needle stylet was then removed and a 20ml syringe was attached to the needle. Four to Five in and out motions of the needle were made. At the same time, suction was applied via the 20ml syringe. Cytological material obtained was injected into a formalin bottle. At least two punctures were made.

Cell blocks were prepared in the pathology laboratory using agar cell blocks to retain tissue fragments. The agar cell blocks were then subjected to tissue processing sequence. The tissue specimens were reviewed by pathologists. ROSE was not available in our center. After the FNA or biopsy, ultrasound was performed again to look for any disappearance of the lesion under pleura suggesting the presence of immediate pneumothorax. Standard chest X ray was performed 4 hours after the procedure to look for pneumothorax. The patient was monitored hourly for 4 hours in the hospital and would be discharged if no complication arose.

There were 105 patients with 113 episodes of pulmonologist-performed ultrasound guided FNA of lung lesions on the list kept by pulmonologists. CDARS search for radiologist-performed image guided CNB yielded 391 results. Computed Tomography was used as the image guidance in 323 cases. A total of 65 patients with 68 ultrasound guided CNB performed by radiologist were included in the radiologist group as historical cohort.

a) Baseline characteristics and lung lesion profile

There was no statistically significant difference in gender, smoking status and lung lesion size between the two groups. There were statistically significant differences between the two groups in age (p-value = 0.02) and lung lesion location (p-value = 0.049). The mean age was 68.7 and 64.4 in pulmonologist group and radiologist group respectively. The lung lesion location was charted in table 1.

b) Diagnostic accuracy and yield

In the pulmonologist group, one episode of FNA was excluded for diagnostic accuracy analysis because the patient died from pneumonia one month after the procedure. In the remaining 112 episodes of FNA, 88 episodes (79%) were regarded as true positive and 2 episodes (2%) were regarded as true negative. There were 22 episodes of false negative (19%) and no false positive. The diagnostic accuracy was 80.4% (95% CI 72-86.7%). The sensitivity is 80% (95% CI 71.6%-86.4%) and negative predictive value is 8.3% (95% CI 2.3%-25.8%). The specificity and positive predictive value were both 100% (table 2).

Among the 68 episodes of CNB in radiologist group, 57 episodes (84%) were regarded as true positive and 2 episodes (3%) were regarded as true negative. There were 9 episodes of false negative (13%) and no false positive. The diagnostic accuracy was 86.8% (95% CI 76.7-92.9%). The sensitivity was 86.4% (95% CI 76.1%-92.7%) and negative predictive value was 18.2% (95% CI 5.1%-47.7%). The specificity and positive predictive value were both 100% (table 2).

The diagnostic accuracy (80.4% vs 86.8%, p-value = 0.27), sensitivity (80% vs 86.4%, p-value = 0.283) and negative predictive value (8.3% vs 18.2%, p-value = 0.575) of pulmonologist and radiologist group were comparable. The results are summarized in table 2.

c) Complication rates

The overall complication rates for pulmonologist group and radiologist group were 5.3% and 7.4% respectively (p-value = 0.749). One patient in each group who developed pneumothorax after the procedure required chest drain insertion while the other patients were managed conservatively with oxygen supplement.

d) The diagnoses

The diagnoses by FNA or CNB were categorized as primary lung cancers (119, 65.7%), metastases (6, 3.3%), hematological malignancies (7, 3.9%), malignancy not otherwise specified (3, 1.7%), infection (8, 4.4%), benign lesions (3, 1.7%) and non-diagnostic (35, 19.3%) (table 3).

e) Predictive factors

In pulmonologist group, univariable regression (with ridge penalty) showed that lesion location of upper lobe was predictive of diagnostic accuracy of ultrasound guided FNA (OR = 5.32, 95% CI 1.78-15.87, p-value = 0.002) (table 4). Combining pulmonologist and radiologist group, univariable regression showed that both lesion size (OR = 1.17, 95% CI 1.0-1.38, p-value = 0.048) and lesion location of upper lobe (OR = 4.96, 95% CI 1.97-12.48, p-value < 0.001) were predictive of diagnostic accuracy of ultrasound guided percutaneous tissue sampling (table 5). Multivariable regression showed that only lesion location of upper lobe (OR = 4.42, 95% CI 1.73-11.28, p-value = 0.002) was predictive of diagnostic accuracy of ultrasound guided percutaneous tissue sampling (table 5).

Our study showed that pulmonologist-performed ultrasound guided FNA of lung lesions had a diagnostic accuracy of 80.4% and a complication rate of 5.3%, which were comparable to a historical cohort of radiologist performed ultrasound guided core needle biopsy of lung lesions in the same hospital (86.8% and 7.4% respectively) (table 2).

a) Diagnostic accuracy in the literature

Pulmonologists in different countries performed ultrasound guided percutaneous tissue sampling with diagnostic yield ranging from 72%(9) to 95%(10) for fine needle aspiration(9-12) and 83%(13) to 93.4%(14) for core needle biopsy(6, 12-14). The diagnostic yield is 70.9%(7) for case series consisting mostly of lung lesion located in the parenchyma. Our study showed a diagnostic accuracy of 80.4%, which was comparable to that in the literatures. The diagnostic yield was similar for biopsy performed by pulmonologists and radiologists in the literature(12). The site of biopsy is not limited to peripheral lung and pleura, but also includes other intrathoracic areas such as rib, mediastinum and chest wall in the literatures(7, 12, 15). From our experience, FNA of rib, cervical lymph node and mediastinal lesions was also feasible but not included in this study.

b) Complication rate in the literature

Complications of ultrasound guided FNA of peripheral lung lesions include pneumothorax and hemoptysis. The overall complication rate was quoted between 2.7% and 15% in the literatures (7, 11-14). The most common complication is pneumothorax, rated from 0% to 7.9%(6, 7, 13, 14). Chest drain insertion and hospital admission are rarely needed for these complications(11). The complication rate of 5.3% in the pulmonologist group in our study highlights the procedure safety.

c) Predictive factors of diagnostic yield

A large study involving more than 9000 patients in Korea on percutaneous image guided biopsy of lung lesions showed that the risk factors for diagnostic failure are lesion size smaller than 2cm, subsolid lesions, FNA without CNB, final diagnosis of benign lesions and lymphoma(16). While specifically for ultrasound guided FNA, factors associated with diagnostic success include suspicion of malignancy, increased size and lack of pleural sliding suggestive of pleural adhesion(11).

In our series, larger lesion size and upper lobe location increased the diagnostic yield (table 4 and table 5). Lesion size is a well-established factor for diagnostic yield. A lesion size of at least 2 cm is appropriate for USG guided FNA. Our study is the first to identify that upper lobe location is predictive of diagnostic accuracy. There are two possible reasons for a higher diagnostic rate in the upper lobe than other lobes. Firstly, the magnitude of lung tumour movement during respiration is larger in the lower lobe than the upper lobe(17, 18). As tissue sampling takes around 10 seconds and patients often cannot hold their breath throughout, lesion movement with respiration may lead to missed target. Secondly, lung lesions in the upper lobe are more likely to be malignant than those in other lobes(19). Upper lobe location is a cancer predictor in lung nodules detected on screening CT(20), which was validated in prediction model such as Brock University cancer prediction equation. The diagnostic accuracy of malignant lesion is much higher than benign lesion by FNA(21). Therefore, lesion location of upper lobe is a positive predictive factor of diagnostic accuracy by FNA.

d) Strengths

Our study had several strengths. Firstly, this is the first local study demonstrating that pulmonologist-performed ultrasound guided FNA of lung lesion has comparable diagnostic accuracy to that of the literature. Secondly, we demonstrated a comparable diagnostic accuracy of FNA and CNB. Thirdly, a low complication rate of the procedure was recorded, ensuring patient safety.

e) Limitations

This study has several limitations. Firstly, the study is a single centre retrospective study comparing the pulmonologist group with the radiologist group which was a historical cohort. Chronology bias existed. Advancement in imaging technology, equipment and pathological diagnostic technique may favour the pulmonologist group. Secondly, there was selection bias in both the pulmonologist group and radiologist group. Difficult cases such as small lesions might be referred by the pulmonologists to the radiologists for CT guided biopsy. Radiologists had variable preference on the choice between CT and ultrasound guidance. Thirdly, there were statistically significant differences in the demographics between the pulmonologist group and the radiologist group in age and lung lesion location. The pulmonologist group cohort was older than that of the radiologist group. We postulated that ultrasound guided FNA was a readily accessible bedside procedure and thus was more acceptable to the elderly, when compared with a scheduled CNB by radiologists. The p-value for lesion location difference is 0.049, which is only marginally significant. Fourthly, there were variations in the equipment, technique, operators’ experience and peri-procedure practices. The procedures were performed by different operators with various experience in this study.

Ultrasound guided FNA of peripheral lung lesions by pulmonologists is an accurate and safe procedure. Lung lesion of upper lobe location is predictive of diagnostic success. Pulmonologists should be encouraged to perform this procedure for making early diagnoses.

ALK = Anaplastic Lymphoma Kinase

CDARS = Clinical Data Analysis and Reporting System

CI = Confidence Interval

CNB = Core Needle Biopsy

CT = Computed Tomography

Endobronchial Ultrasound = EBUS

EGFR = Epidermal Growth Factor Receptor

FNA = Fine Needle Aspiration

ICD = International Code of Disease

OR = Odds Ratio

PDL1 = Programmed cell Death Ligand 1

ROSE = Rapid On-Site Evaluation

Transbronchial Needle Aspiration = TBNA

Ethics approval and consent to participate:

This study has not been presented, published or posted online before. The study protocol was approved by the Kowloon West Cluster Research Ethics Committee of the Hospital Authority, Hong Kong with the reference number of KW/EX-21-128(163-06). In view of the retrospective nature, informed consent was waived by the Kowloon West Cluster Research Ethics Committee of the Hospital Authority, Hong Kong. All methods were carried out in accordance with the regulations of the committee. The patient was treated in accordance with the Declaration of Helsinki.

Consent for publication:

Not applicable

Availability of data and materials

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Competing Interest:

All authors have disclosed no conflicts of interest.

Funding:

This case report received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Author contributions:
Concept or design: CT Kwok, YC Yeung.
Acquisition of data: CT Kwok
Analysis or interpretation of data: CT Kwok
Drafting of the manuscript: CT Kwok
Critical revision of the manuscript for important intellectual content: YC Yeung, YH Chan, MY Ho
All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Acknowledgements:

Respiratory team members of the Princess Margaret Hospital* performed ultrasound guided fine needle aspiration of peripheral lung masses. The service was led by Dr Yeung Yiu Cheong. Dr Yu Wai Cho, Dr Yeung Yiu Cheong, Dr Chan Yu Hong and Dr Ho Man Ying reviewed the final report. Ms Ellen Yu, statistician of the Princess Margaret Hospital, aided in the statistical part of the study.

Gould MK, Tang T, Liu IL, Lee J, Zheng C, Danforth KN, et al. Recent Trends in the Identification of Incidental Pulmonary Nodules. Am J Respir Crit Care Med. 2015;192(10):1208-14.
Detterbeck FC, Lewis SZ, Diekemper R, Addrizzo-Harris D, Alberts WM. Executive Summary: Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2013;143(5 Suppl):7S-37S.
Yung RC. Tissue diagnosis of suspected lung cancer: selecting between bronchoscopy, transthoracic needle aspiration, and resectional biopsy. Respir Care Clin N Am. 2003;9(1):51-76.
Koh DM, Burke S, Davies N, Padley SP. Transthoracic US of the chest: clinical uses and applications. Radiographics. 2002;22(1):e1.
Targhetta R, Bourgeois JM, Marty-Double C, Coste E, Proust A, Balmes P, et al. Peripheral pulmonary lesions: ultrasonic features and ultrasonically guided fine needle aspiration biopsy. J Ultrasound Med. 1993;12(7):369-74.
Diacon AH, Schuurmans MM, Theron J, Schubert PT, Wright CA, Bolliger CT. Safety and yield of ultrasound-assisted transthoracic biopsy performed by pulmonologists. Respiration. 2004;71(5):519-22.
Rodriguez EF, Pastorello R, Osmani L, Hopkins M, Kryatova M, Kawamoto S, et al. Ultrasound-Guided Transthoracic Fine-Needle Aspiration: A Reliable Tool in Diagnosis and Molecular Profiling of Lung Masses. Acta Cytol. 2020;64(3):208-15.
Koegelenberg CF, Bolliger CT, Plekker D, Wright CA, Brundyn K, Louw M, et al. Diagnostic yield and safety of ultrasound-assisted biopsies in superior vena cava syndrome. Eur Respir J. 2009;33(6):1389-95.
Knox D, Halligan K. Case series of trans-thoracic nodule aspirate performed by interventional pulmonologists. Respir Med Case Rep. 2021;32:101362.
Cozzolino I, Ronchi A, Messina G, Montella M, Morgillo F, Vicidomini G, et al. Adequacy of Cytologic Samples by Ultrasound-Guided Percutaneous Transthoracic Fine-Needle Aspiration Cytology of Peripheral Pulmonary Nodules for Morphologic Diagnosis and Molecular Evaluations: Comparison With Computed Tomography-Guided Percutaneous Transthoracic Fine-Needle Aspiration Cytology. Arch Pathol Lab Med. 2020;144(3):361-9.
Meena N, Bartter T. Ultrasound-guided Percutaneous Needle Aspiration by Pulmonologists: A Study of Factors With Impact on Procedural Yield and Complications. J Bronchology Interv Pulmonol. 2015;22(3):204-8.
Portela-Oliveira E, Souza CA, Gupta A, Bayanati H, Inacio J, Rakhra K. Ultrasound-guided percutaneous biopsy of thoracic lesions: high diagnostic yield and low complication rate. Clin Radiol. 2021;76(4):281-6.
Huang W, Ye J, Qiu Y, Peng W, Lan N, Huang T, et al. Ultrasound-Guided Percutaneous Core Needle Biopsy of Peripheral Pulmonary Nodules </= 2 cm: Diagnostic Performance, Safety and Influence Factors. Front Oncol. 2021;11:671884.
Yamamoto N, Watanabe T, Yamada K, Nakai T, Suzumura T, Sakagami K, et al. Efficacy and safety of ultrasound (US) guided percutaneous needle biopsy for peripheral lung or pleural lesion: comparison with computed tomography (CT) guided needle biopsy. J Thorac Dis. 2019;11(3):936-43.
Abraham AE, Suresh PK, Sridevi HB, Sahu KK, Adiga D, Minal J, et al. Image-guided Fine Needle Aspiration Cytology of Intrathoracic Lesions. J Cytol. 2019;36(2):106-10.
Lee KH, Lim KY, Suh YJ, Hur J, Han DH, Kang MJ, et al. Diagnostic Accuracy of Percutaneous Transthoracic Needle Lung Biopsies: A Multicenter Study. Korean J Radiol. 2019;20(8):1300-10.
Knybel L, Cvek J, Molenda L, Stieberova N, Feltl D. Analysis of Lung Tumor Motion in a Large Sample: Patterns and Factors Influencing Precise Delineation of Internal Target Volume. Int J Radiat Oncol Biol Phys. 2016;96(4):751-8.
Shimizu S, Shirato H, Kagei K, Nishioka T, Bo X, Dosaka-Akita H, et al. Impact of respiratory movement on the computed tomographic images of small lung tumors in three-dimensional (3D) radiotherapy. Int J Radiat Oncol Biol Phys. 2000;46(5):1127-33.
Swensen SJ, Silverstein MD, Ilstrup DM, Schleck CD, Edell ES. The probability of malignancy in solitary pulmonary nodules. Application to small radiologically indeterminate nodules. Arch Intern Med. 1997;157(8):849-55.
McWilliams A, Tammemagi MC, Mayo JR, Roberts H, Liu G, Soghrati K, et al. Probability of cancer in pulmonary nodules detected on first screening CT. N Engl J Med. 2013;369(10):910-9.
Greif J, Marmor S, Schwarz Y, Staroselsky AN. Percutaneous core needle biopsy vs. fine needle aspiration in diagnosing benign lung lesions. Acta Cytol. 1999;43(5):756-60.

Table 1. Baseline characteristics and lung lesion profile of patients

		Pulmonologist group (n=113)	Radiologist group (n=68)	P-value
Age, year, mean±SD [range]		68.7±12 [26-94]	64.4±12.6 [31-84]	0.020
Gender (male:female)		85:28	55:13	0.378
Smoking status, n(%)				0.558
	Non-smoker	35(31)	22(32.5)
	Ex-smoker	33(29)	24(35)
	Smoker	45(40)	22(32.5)
Lung lesion size, centimeter, mean±SD [range]		6.5±2.9 [1.7-19]	6.7±3.2 [2-17]	0.687
Lung lesion location, n(%)				0.049
	Right upper lobe	40(35)	20(29.4)
	Right middle lobe	7(7)	3(4.4)
	Right lower lobe	26(23)	16(23.5)
	Left upper lobe	17(15)	17(25)
	Left lingula lobe	0	4(5.9)
	Left lower lobe	23(20)	8(11.8)

Table 2. Diagnostic accuracy and yield

	Pulmonologist group (n=112)	Radiologist group (n=68)	p-value
Accuracy (95%CI)	80.4% (72.0%-86.7%)	86.8% (76.7%-92.9%)	0.270
Sensitivity (95% CI)	80% (71.6%-86.4%)	86.4% (76.1%-92.7%)	0.283
Specificity (95% CI)	100% (34.2%-100%)	100% (34.2%-100%)	1
Positive predictive value (95% CI)	100% (95.8%-100%)	100% (93.7%-100%)	1
Negative predictive value (95% CI)	8.3% (2.3%-25.8%)	18.2% (5.1%-47.7%)	0.575

Table 3. Diagnoses by fine needle aspiration or core needle biopsy (total n=181)

		Pulmonologist group (n=113)	Radiologist group (n=68)
Lung cancers
	Adenocarcinoma	44 (39%)	18 (26%)
	Squamous cell carcinoma	14 (12%)	14 (21%)
	Non-small cell carcinoma	10 (9%)	3 (4%)
	Small cell carcinoma	6 (5%)	3 (4%)
	SMARCA4-deficient malignant tumor	1 (1%)	0
	Sarcomatoid carcinoma	0	2 (3%)
	Lymphoepithelioma-like carcinoma	2 (2%)	1 (1%)
	Adenosquamous carcinoma	1 (1%)	0
Metastasis		1 (1%)	5 (7%)
Hematological malignancies
	Lymphoma	3 (3%)	2 (3%)
	Leukemic infiltrate	1 (1%)	1 (1%)
Malignancy not otherwise specified
	Malignant cells	1 (1%)	0
	Carcinoma	1 (1%)	0
	Spindle cell lesion	0	1 (1%)
Infection
	Aspergillosis	1 (1%)	2 (3%)
	Cryptococcal infection	0	1 (1%)
	Tuberculosis	3 (3%)	1 (1%)
Benign lesions
	Fibrous tumour	0	2 (3%)
	Neurilemmoma	0	1 (1%)
Non-diagnostic
	Acute inflammation	2 (2%)	0
	Necrosis	3 (3%)	1 (1%)
	Suspicious cells	3 (3%)	1 (1%)
	Atypical cells	2 (2%)	1 (1%)
	Quantity insufficiency	7 (6%)	1 (1%)
	Negative	7 (6%)	7 (10%)

Table 4. Predictive factor of diagnostic accuracy in pulmonologist group

	Univariable regression
	OR_unadj (95% CI)		p
Lesion size	1.12	(0.93 - 1.34)	0.230
Lesion location^a
Lower (ref.)	1		—
Middle/ lingular	44.67	(0.18 - 11240.83)	0.173
Upper	5.32	(1.78 - 15.87)	0.002
Age	1.01	(0.98 - 1.05)	0.475
Male	1.53	(0.55 - 4.26)	0.412
Smoking status
Never (ref.)	1		—
Ex	1.28	(0.39 - 4.21)	0.680
Current	1.37	(0.46 - 4.11)	0.574
^a With ridge penalty

Table 5. Predictive factors of diagnostic accuracy

		Univariable regression			Multivariable regression
		OR_unadj (95% CI)		p	OR_adj (95% CI)		p
Lesion size, cm		1.17	(1.00 - 1.38)	0.048	1.12	(0.95 - 1.31)	0.176
Lesion location
	Lower (ref.)	1		—	1		—
	Middle/ lingular	1.48	(0.38 - 5.85)	0.575	1.39	(0.35 - 5.56)	0.638
	Upper	4.96	(1.97 - 12.48)	<0.001	4.42	(1.73 - 11.28)	0.002
Age, year		1.01	(0.98 - 1.04)	0.475	—		—
Male		1.81	(0.77 - 4.24)	0.171	—		—
Smoking status
	Never (ref.)	1		—
	Ex	1.09	(0.44 - 2.73)	0.852	—		—
	Current	1.97	(0.74 - 5.21)	0.174	—		—

No competing interests reported.

Pulmonologist performed ultrasound guided fine needle aspiration of lung lesions

Status:

Version 1

Abstract

Figures

Introduction

Objectives

Primary Objective:

Secondary Objectives:

Methods

a) Subjects

b) Diagnostic category

c) Study design

d) Statistical analysis

e) The ultrasound guided fine needle aspiration

Results

a) Baseline characteristics and lung lesion profile

b) Diagnostic accuracy and yield

c) Complication rates

d) The diagnoses

e) Predictive factors

Discussions

a) Diagnostic accuracy in the literature

b) Complication rate in the literature

c) Predictive factors of diagnostic yield

d) Strengths

e) Limitations

Conclusion

Abbreviations

Declarations

References

Tables

Additional Declarations

Status:

Version 1