Skip to main content
Download PDF

Impact of radiomics features, pulmonary emphysema score and muscle mass on the rate of pneumothorax and chest tube insertion in CT-guided lung biopsies

Respiratory Research volume 25, Article number: 320 (2024) Cite this article

Abstract

Iatrogenic pneumothorax is a relevant complication of computed tomography (CT)-guided percutaneous lung biopsy. The aim of the present study was to analyze the prognostic significance of texture analysis, emphysema score and muscle mass derived from CT-imaging to predict postinterventional pneumothorax after CT-guided lung biopsy. Consecutive patients undergoing CT-guided percutaneous lung biopsy between 2012 and 2021 were analyzed. Multivariate logistic regression analysis included clinical risk factors and CT-imaging features to detect associations with pneumothorax development. Overall, 479 patients (178 females, mean age 65 ± 11.7 years) underwent CT-guided percutaneous lung biopsy of which 180 patients (37.5%) developed pneumothorax including 55 patients (11.5%) requiring chest tube placement. Risk factors associated with pneumothorax were chronic-obstructive pulmonary disease (COPD) (p = 0.03), age (p = 0.02), total lung capacity (p < 0.01) and residual volume (p = 0.01) as well as interventional parameters needle length inside the lung (p < 0.001), target lesion attached to pleura (p = 0.04), and intervention duration (p < 0.001). The combined model demonstrated a prediction accuracy of the occurrence of pneumothorax with an AUC of 0.78 [95%CI: 0.70–0.86] with a resulting sensitivity 0.80 and a specificity of 0.66. In conclusion, radiomics features of the target lesion and the lung lobe CT-emphysema score are predictive for the occurrence of pneumothorax and need for chest insertion after CT-guided lung biopsy.

Introduction

Computed tomography (CT)-guided biopsy of unclear/suspicious pulmonary nodules is routinely used to obtain tissue for histopathology assessment. In comparison with bronchoscopy biopsy, CT-guided biopsies are considered cost effective and yield a higher diagnostic accuracy in regard of the assessment of peripheral located pulmonary lesions or nodules [1,2,3].

In comparison with surgical approach, percutaneous CT-guided lung biopsies can be considered less invasive and does not require anesthesia [2, 4].

However, CT-guided lung biopsies harbor also interventional complications and risks [3, 4]. Additionally, to the known complications of other CT-guided percutaneous interventions comprising infectious seeding, hematoma and tissue damage, lung biopsies have specific complications, especially the incidence of pneumothorax [3].

Pneumothorax is considered as the most common complication of lung biopsies with yet high and variable reported incidences. According to the literature, the frequency ranges from 9 to 54% throughout the different study populations [5]. A comprehensive analysis of total 9,783 CT-guided lung biopsies reported a frequency of 35% [4]. Notably, the amount of cases with the need of chest tube placement ranged from 2 to 15% of cases [6].

Various studies investigated factors that might influence the amount of postinterventional pneumothorax and consequently therapeutic release with chest tube insertion, including the risk factor pulmonary emphysema [4,5,6,7,8,9,10]. Among these, lung emphysema was predominantly assessed visually by the reader, which can be considered subjective and could be influenced by bias.

During recent years, quantitative CT imaging markers are a reliable method to quantify emphysematous lung areas as low attenuation areas in thoracic CT images using a dedicated software tool [5].

To date, only few systematic studies, each only investigating small patient samples, have investigated quantitative imaging methods in these patients. Despite some evidence in favor of quantitative emphysema score as a risk factor for occurrence of pneumothorax [5], some data disapprove it as a risk factor [11].

In the search of better methods to predict pneumothorax in patients at risk following CT-guided lung biopsy, texture analysis is a research field to provide quantitative imaging biomarkers using radiological images [12,13,14,15].

Various studies have investigated the clinical benefit of the use of CT texture analysis, most commonly in the research field of oncology [12,13,14,15]. Using this approach, texture features can provide imaging information, which provide more insight into microstructure of tissue. It is unclear so far, whether texture analysis could aid to predict pneumothorax risk undergoing CT-guided lung biopsy. In the light of the upcoming increase in low-dose CT studies within the pan-European lung cancer screening programmes for patients at risk of lung cancer, CT-guided lung biopsies will represent an adequate procedure of peripheral lesion in certain patients [16]. The major risk developing pneumothorax for those individuals following transthoracic lung biopsy is mainly caused by a relevant smoking history leading to airway limitation and irreversible parenchymal lung tissue destruction, and thus chronic obstructive pulmonary disease (COPD) and emphysema [17].

The purpose of the present analysis was to elucidate the rate of pneumothorax following CT-guided lung biopsy and to identify risk factors with clinical features and quantitative CT imaging comprising CT lung emphysema score and texture analysis of the target lesion.

Materials and methods

Patient acquisition

The present retrospective analysis was approved by the institutional review board (University of Leipzig, register no. 344–2007, Ethics Committee, University of Leipzig, Leipzig, Germany). All patients undergoing CT-guided lung biopsy were retrospectively assessed between the years 2012 and 2021. COPD diagnosis was documented according to patient’s medical history.

Clinical and respiratory parameters

Comorbidities were extracted from the medical records comprising diabetes mellitus, arterial hypertension and prior malignancy. COPD diagnosis was documented according to patient’s medical history. History of smoking and the amount of pack years (20 cigarettes per pack per day over the years) were extracted from the records.

The COPD was classified accordingly to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) [17]. The pulmonary function test (PFT) measurements included spirometry and body plethysmography and were performed by using the MasterScreen-Body by JAEGER GmbH & Co. KG according to current practice guidelines on technical standards [18, 19] and included forced vital capacity (FVC), forced expiratory volume in 1s (FEV1), its ratio (FEV1 /FVC), total airway resistance (Rtot), total lung capacity (TLC), residual volume (RV) and its ratio (RV/TLC) as well as diffusion capacity for carbon monoxide after a single breath (DLCO SB) and DLCO SB referenced to alveolar volume (DLCO/VA) were assessed.

Blood gas analyses (BGA) were performed on capillary blood samples analyzed. BGA included arterial oxygen tension (pO2), arterial carbon dioxide tension (pCO2) and arterial oxygen saturation (SaO2) were performed at the room temperature. PH values, and the alveolar-arterial gradient of oxygen (AaDO2) were calculated [20].

Procedure of the CT-guided biopsy

The patients gave written informed consent one day before the CT-guided biopsy. The biopsy was only indicated without increased tendency to bleed indicated when platelet count was at least 50.000/mm3, partial thromboplastin time ≤ 1.5 times and prothrombin time > 50%. All interventions were performed with the same CT scanner (16-slice CT scanner, Brilliance Big Bore, Philips, Hamburg, Germany). Typical CT parameters were set: 100 kVp; 125 mAs; slice thickness,1 mm; pitch, 0.9).

All biopsies investigated in this study were performed by interventional trained radiologists with at least 2 years of general experience in interventional radiology. The procedure plan, comprising the position and the needle pathway were planned using the last CT images according to clinical routine. Basic principles of the intervention comprised assessing an angle vertical to the parietal pleura, avoiding the lung fissures, large bronchovascular structures, minimizing the pathway of the needle. At last, the passage was planned at the upper part of the ribs to reduce potentially of hematoma [21].

The procedure started with disinfection of the skin and local anesthesia with 10 ml lidocaine 1% (Xylocitin, Jenapharm, Germany) were performed. In all instances, a coaxial 18-gauge biopsy system of a 2 cm throw length needle was used (Bard Mission, Bard Medical, Covington, GA 30014, USA or Biopince, Argon Medical Devices, Athens, TX 75751, USA). During the procedures, CT images were used to validate the accurate localization of the needle tip to reach the target nodule correctly. After removal of the biopsy needle, CT images over the whole lung were acquired to detect post-interventional complications. Patients needed to rest without eating or drinking up to two hours after the procedure to minimize the occurrence of post-biopsy pneumothorax. A chest radiograph was acquired two hours after the biopsy to diagnose complications especially the occurrence of pneumothorax. Patients with pneumothorax on immediate post-biopsy CT images were labelled as “instant pneumothorax”. Cases were classified as “delayed pneumothorax”, when a new pneumothorax was identified on the following radiograph. Cases with a pneumothorax of > 10 mm hem width and/or with clinical symptoms of shortness of breath, raised heart rate and declining oxygen saturation received chest tube insertion.

CT emphysema score

In the literature, a Hounsfield unit (HU) cut-off value of below − 950 is used to discriminate emphysematous to physiological parenchyma of the lung [5, 22]. For this study, the software package from Philips (Intellispace portal, version 11; Philips, Philips Health System, Hamburg, Germany) was used. Areas of the segmented lung parenchyma − 950 HU were automatically identified and utilized to obtain the total emphysema areas and consecutively provide a score. In short, the total score is calculated as the number of voxels below − 950 HU/voxels of the lung. A description can be found in another study [5]. Moreover, the same measurement was performed for the lung lobe of the target lesion to provide a measurement of the surrounding lung parenchyma.

Skeletal muscle area

A representative CT-slice at the level of the 12th thoracic vertebral body was used to calculate the skeletal muscle area. The area containing the paravertebral and intercostal muscles was measured on this slice using the freely available software ImageJ 1.48v (National Institutes of Health Image program). A Hounsfield-unit threshold value of -29 to 150 HU was applied to semiautomatically identify the muscle mass area giving the total skeletal muscle area (SMA) of this slice (cm2) [23].

Texture analysis

CT images were further analyzed with the dedicated software MaZda (version 4.7, available at http://www.eletel.p.lodz.pl/mazda/) [24, 25]. A region of interest (ROI) was drawn on the largest slice of the target lesion to calculate the texture features. The ROI was drawn clearly within the pulmonary lesion with two mm distance to the adjacent lung parenchyma. Texture analysis measurements were blindly carried out to the bioptic results by one radiologist (J.L.) with 4 years of general radiological experience. For each ROI, gray-level (µ) normalization was utilized to µ ± 3 standard deviations to reduce the influence of contrast and brightness variation on the texture features, as performed previously [26, 27].

Several different texture features were extracted for each patient including histogram parameters, second order texture features of different groups comprising (co-occurrence matrix run-length matrix, absolute gradient, autoregressive model (theta 1 to 4, sigma), and wavelet transform features. Altogether, 279 texture features were calculated in all lesions.

Statistical analysis

For this study, statistical analysis was performed with SPSS (IBM, Version 25.0; Armonk, NY, USA). Collected results were primarily assessed with descriptive statistics. Group differences were further analyzed with Mann-Whitney-U test and Fisher’s exact test. Uni- and multivariate logistic regression analysis was carried out to elucidate possible predictors for the occurrence of the postinterventional pneumothorax, given as odds ratio (OR) with 95% confidence interval (CI). Receiver operating characteristics-curve (ROC) with area under the curve (AUC) analyses were used to investigate diagnostic accuracy for the investigated parameters. P-values below 0.05 were used to indicate statistical significance.

Results

Overall, 479 patients (178 female patients, 37.2%) were analyzed. The mean age was 65 ± 11.7 years, ranging from 25 to 91 years. Representative images of the patient sample are displayed in Fig. 1.

Fig. 1
figure 1

Representative images of the patient sample. First row: segmentation of the lung in order to achieve automated emphysema scoring. Second row: representative slice of the CT Thorax planning scan demonstrating the biopsied lung lesion. Third row: drawn region of the interest (in red) of the biopsied lung lesion. First line: 71-year-old male patient, right-elevated position, no pneumothorax after biopsy, biopsy successful (non-malignant lesion). Second line: 67-year-old female patient, right-lateral position, pneumothorax after biopsy, biopsy successful (metastatic lesion)

Full size image

Influences on the occurrence of postinterventional pneumothorax

A total of 180 patients had a postinterventional pneumothorax (37.6%). An overview of the clinical features including group comparisons is given by Table 1.

Table 1 Overview of the clinical features of the patient sample
Full size table

Statistically significant differences were identified between the patient groups “no pneumothorax” and “any pneumothorax” for the clinical parameters age (64.95 ± 12.10 vs. 67.56 ± 10.91 years, (OR 1.02; [95% CI 1.00-1.04]; p = 0.02), COPD (OR 1.67, [95% CI 1.05–2.67], p = 0.03), the total lung capacity (OR 1.02; 95% [CI 0.99–1.05], p < 0.01) and the residual lung volume (OR 0.99; 95% [CI 0.98-1.00], p = 0.01).

The inserted needle length also showed statistically significant difference between appearance of a pneumothorax or no pneumothorax (3.11 ± 2.08 cm vs. 3.96 ± 2.34 cm; OR 1.19, [95% CI 1.04–1.25], p < 0.001). Different ways of positioning the patient showed statistical significance: right lateral (OR 1.09, 95% [CI 0.58–2.08], p = 0.03), supine (OR 1.19, [95% CI 0.82–1.72], p > 0.01) and prone position (OR 1.10, [95% CI 0.71–1.70], p = 0.01) raised the risk of a pneumothorax.

Opposing to that, lesions attachment to the pleura (OR 0.68, [95% CI 0.47–0.99], p = 0.04) and greater lesion size (7.88 ± 13.92 mm vs. 4.58 ± 7.04 mm, OR 0.96, [95% CI 0.94–0.99], p = 0.04) lowered the risk of pneumothorax. The duration of the intervention differed statistically significant between those without pneumothorax (0:15:21.54 ± 0:08:28.39) and those with pneumothorax (0:18:17.55 ± 0:09:17.60), p < 0.001.

Several quantitative CT features were also statistically significant different between the patient groups as provided by Table 2. Total emphysema score was higher in patients with pneumothorax (14.50 ± 11.67 vs. 12.13 ± 10.91, p = 0.04, OR 1.02 [95% CI 1.00–1.04]).

Table 2 Quantitative CT features between the patient groups with postinterventional pneumothorax and without
Full size table

Also, the low attenuation volume of the biopsied lobe was higher in patients with pneumothorax (168.42 ± 121.66 compared to 213.07 ± 146.87, p = 0.01, OR 1.003 [95% CI 1.001–1.004]). Furthermore, the total volume of the biopsied lobe (890.57 ± 429.71 vs. 970.09 ± 404.17, p = 0.04, OR 0.99 [95% CI 0.97–1.00]) and the low attenuation volume ratio (17.87 ± 8.75 vs. 20.42 ± 8.99, p = 0.01, OR 0.91 [95% CI 0.82–1.00]) were significantly different between the groups.

The texture features “S(1,0)SumAverg” (66.11 ± 0.79 for no pneumothorax vs. 66.39 ± 0.68 for pneumothorax, p < 0.001, OR 1.26 [0.75–2.11]) and “GrMean” (2.85 ± 0.74 for no pneumothorax vs. 2.65 ± 0.63 for any pneumothorax, OR 1.02, 0.31–3.33, p = 0.03) showed noticeable differences between both categories.

Multivariate logistic regression was carried out to elucidate the relationships between the statistically significant clinical (n = 9), imaging (n = 4) and CT texture features (n = 7) and the occurrence of any pneumothorax. The combined model demonstrated a prediction accuracy of the occurrence of pneumothorax with an AUC of 0.78 [95%CI: 0.70–0.86] with a resulting sensitivity 0.80 and a specificity of 0.66. Using only clinical features yielded an AUC of 0.73 [95% CI: 0.65–0.81], while the sole texture and imaging feature model resulted in an AUC of 0.64 [95% CI: 0.54–0.73]. The corresponding graphs are displayed in Fig. 2.

Fig. 2
figure 2

Receiver operating characteristics-curve (ROC) of the prediction of the appearance of a pneumothorax in multivariate logistic regression analysis, clinical model (beige), texture model (green) and combined model (blue). The combined mode achieved an AUC of 0.78 [95%CI: 0.70–0.86], the clinical model achieved an AUC of 0.73 [95% CI: 0.65–0.81], while the texture model an AUC of 0.64 [95% CI: 0.54–0.73]

Full size image

Subgroup analysis of chest tube placement

In 55 of the 180 cases (30.6%) with pneumothorax, there was the need of a chest tube placement after biopsy. The clinical features for these patient groups are shown in Table 3. The statistically significant clinical parameters were the anamnestic presence of COPD (OR 2.86 [95% CI 1.53–5.33], p = 0.01), the forced vital capacity (2.62 ± 0.98 for no chest tube necessity vs. 2.96 ± 0.79, OR 1.24 [95% CI 0.76-2.00] p = 0.02) and the forced vital capacity in percent (78.47 ± 20.24 for no chest tube necessity vs. 87.32 ± 20.28, OR 1.02 [95% CI 0.99–1.04], p = 0.02).

Table 3 Clinical features for the patient groups “necessity of chest tube insertion” and “no necessity of chest tube insertion”
Full size table

Regarding quantitative CT features, n = 18 parameters were statistically significant different between these groups. The highest significance demonstrated the histogram feature Perc,01% (65.92 ± 34.59 vs. 79.63 ± 38.98, OR 1.01 [95%CI 1.00-1.02], p = 0.01) and the second order feature S(1,-1)Correlat (0.64 ± 0.15 vs. 0.56 ± 0.21, OR 0.16 [95%CI 0.02–1.24], p = 0.02).

Multivariate logistic regression was used to analyze the relationship between the significant clinical features (n = 3) and the above mentioned highest significant texture features (n = 2) with the necessity of a chest tube. The combined model demonstrated a prediction accuracy of the occurrence of pneumothorax with an AUC of 0.70 [95% CI: 0.60–0.79] resulting in an optimized sensitivity of 0.72 with a specificity of 0.55. The sole texture feature model yielded an AUC of 0.66 [95% CI: 0.56–0.76], while the sole clinical model resulted in an AUC of 0.65 [95% CI: 0.55–0.75]. The corresponding graphs are displayed in Fig. 3.

Fig. 3
figure 3

Receiver operating characteristics-curve (ROC) of the prediction of the necessity of a chest tube in the patient subgroup of appearance of any pneumothorax, clinical model (beige), texture model (green), combined model (blue). The combined model achieved an AUC of 0.70 [95% CI: 0.60–0.79], the clinical model achieved an AUC of 0.65 [95% CI: 0.55–0.75], while the texture model achieved an AUC of 0.66 [95% CI: 0.56–0.76]

Full size image

Discussion

The present study elucidated parameters to predict pneumothorax and the necessity of chest tube insertion after CT-guided lung biopsy by means of clinical features and quantitative CT image parameters based on a large patient sample.

The frequency of pneumothorax following CT-guided lung biopsy reported herein (30.6%) was slightly higher compared to the pooled frequency of 25.3% reported in a meta-analysis comprising 32 studies [3]. However, this might be explained by the fact that only 18G systems were used, which result in a frequency of pneumothorax compared to 20G systems [3].

The investigated CT parameters comprised CT texture analysis, skeletal muscle mass and quantitative emphysema scores. Further factors of the interventional procedure such as characteristics of the punctuated mass and technical aspects were investigated. Several differences were identified between the patient group with the occurrence of pneumothorax such as age, emphysema scores and established diagnosis of COPD in the medical history.

Also, the technical aspect of the inserted length of the needle as well as positioning of the patient showed statistically significant differences. Indeed, right lateral, supine and prone positioning showed significantly higher rates of pneumothorax as did a longer insertion distance of the biopsy needle.

Interestingly, imaging aspects of the punctured lesions also might have played a role in the risk of a pneumothorax: lesion size as well as attachment to pleura and multiple texture parameters were statistically different between both groups highlighting the influence of these parameters on the interventional planning phase. These characteristics are in good agreement with previous analyses [3, 28, 29].

In fact, bigger lesions with less distance to the pleura happened to develop a periinterventional pneumothorax less commonly. This might be caused by smaller and more deeply located lesions being fundamentally easier missed with subsequent need of correction of the needle position and therefore further manipulation putting stress on the pleura. Adding to this, breathing-caused variability of the lesions position relatively to the needle might also be greater in smaller and more deeply located targets. Finally, also duration of the biopsy procedures showed statistical significance, a factor that might similarly be accounted for by more manipulation of the needle and therefore also at the pleura, increasing the risk of changes in the pleural air pressure levels. Skeletal muscle mass and solidity of the punctuated lesions did not differ statistically significant between occurrence of pneumothorax and no pneumothorax.

Regarding emphysema score, our findings are in line with earlier conducted studies who also found these correlations [3, 11, 30, 31]. Two of these studies also found lesion size and inserted needle length to be associated with the risk of a pneumothorax [11, 30]. Also, higher distance to pleura has already been shown to be a risk factor, which is in line with our results. Furthermore, the literature shows that the solidity of punctuated lesions does not cause an increased risk of development of a pneumothorax, which is also matching our results [32]. Interestingly, we also found a statistically significant higher mean age of the patients that developed a pneumothorax when compared to patients without that complication. This finding is somewhat different to the results of studies that have been earlier conducted [33].

In addition, a recent publication stated the “pleura tail sign”, which is a linear extension of the visceral pleura connecting to the lesion, as a possible predictor of a high risk of developing a pneumothorax [34].

However, the impact of CT texture analysis as a risk factor for development of a pneumothorax was not analyzed previously. We identified that numerous texture features showed statistically significant differences in punctuated lesions with peri- or postprocedural development of a pneumothorax when being compared to lesions being punctuated without subsequent pneumothorax. This might be explainable due to the nature of texture parameters as possible biomarkers reflecting lesion heterogeneity [12,13,14,15] which has been shown to possibly reflect underlying histopathology features and therefore also the relation between the lung lesion and its surrounding lung parenchyma. When conducting multivariate regression analysis and building a prediction model, a combined clinical and texture feature model even yielded a higher AUC than the sole clinical model. The identified accuracy of the current model is in line to earlier conducted studies that also sought to predict the complication of a pneumothorax in lung biopsies [34].

When looking into the necessity of the insertion of a chest tube, we highlighted the factor of existing COPD being of statistical significance. Moreover, CT texture features describing the heterogeneity of the punctuated masses were also of predictive nature. Interestingly, the emphysema scores did not show any statistically significant differences between the subgroups of pneumothorax with necessity of chest tube and pneumothorax without necessity of a chest tube, which could lead to the conclusion that the emphysema score is only predictive of the occurrence of the pneumothorax but not for the severity.

Only a very limited number of studies sought to examine possible associations of imaging and clinical data with the necessity of chest tube insertion after CT-guided biopsy of lung nodules. Notably, most of them did not evaluate biopsies with an 18G system. Some of these studies found that higher emphysema scores are raising the risk of the necessity of insertion of a chest tube [3, 33, 35], while others like our study did not [31]. A key finding of our study is that the CT-defined emphysema score of the lobe of the target lesion can better predict postinterventional pneumothorax compared to the global CT-emphysema score. It seems plausible that the surrounding lung parenchyma is of greater importance as the biopsy needle passes this parenchyma directly. This new interesting finding should be included in further studies.

One merit of the present analysis was the combination of radiomics features of the target lesion, clinical information and emphysema score to comprehensively predict the postinterventional pneumothorax.

The present analysis is the first investigating the importance of CT-based features of sarcopenia in patients undergoing lung biopsy. Notably, the CT defined muscle mass was not an important risk factor for pneumothorax occurrence, albeit it is a risk factor for various complications throughout surgery and interventional radiology [36,37,38]. It remains elusive, whether this fact was confounded by the measurement on the level Th12.

In addition, we found many CT-derived texture features of the punctuated lesions to be significantly different between successful biopsies and biopsies without diagnostic result, highlighting this quantitative CT parameters in further analyses.

The present analysis is not free of limitations. First, it is a retrospective study with known inherent bias. Second, although in all cases an 18G needle was used for the biopsy, the needle length varied slightly due to the localization of the target lesion. Third, there is some clinically inconsistency when indicating a chest tube, it is based upon clinical symptoms, diameter of the pneumothorax and even operator’s preference. In some rare cases, another interventionalist might have decided to place a chest tube or not to place it in a different manner. Forth, all clinical parameters were not available for every case, which is caused by the retrospective study design. As CT-guided biopsy patients are a heterogeneous cohort, which are not investigated by pulmonologists in every case.

Conclusion

Radiomics features of the target lesion and the lung lobe CT-emphysema score are predictive for the occurrence of pneumothorax and need for chest insertion after CT-guided lung biopsy. The identified parameters might aid in clinical decision-making process, and could identify patients at risk for pneumothorax after CT-guided lung biopsy.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

COPD:

Chronic obstructive pulmonary disease

CT:

Computed tomography

HU:

Hounsfield unit

SMA:

Skeletal muscle area

References

  1. Dale CR, Madtes DK, Fan VS, Gorden JA, Veenstra DL, et al. Navigational bronchoscopy with biopsy versus computed tomography-guided biopsy for the diagnosis of a solitary pulmonary nodule: a cost-consequences analysis. J Bronchol Interv Pulmonol. 2012;19:294–303.

    Article  Google Scholar 

  2. Bhatt KM, Tandon YK, Graham R, et al. Electromagnetic navigational bronchoscopy versus CT-guided percutaneous sampling of Peripheral Indeterminate Pulmonary nodules: a Cohort Study. Radiology. 2018;286:1052–61.

    Article  PubMed  Google Scholar 

  3. Heerink WJ, de Bock GH, de Jonge GJ, et al. Complication rates of CT-guided transthoracic lung biopsy: meta-analysis. Eur Radiol. 2017;27:138–48.

    Article  CAS  PubMed  Google Scholar 

  4. Tomiyama N, Yasuhara Y, Nakajima Y, et al. CT-guided needle biopsy of lung lesions: a survey of severe complication based on 9783 biopsies in Japan. Eur J Radiol. 2006;59:60–4.

    Article  PubMed  Google Scholar 

  5. Theilig D, Petschelt D, Mayerhofer A, et al. Impact of quantitative pulmonary emphysema score on the rate of pneumothorax and chest tube insertion in CT-guided lung biopsies. Sci Rep. 2020;10:10978.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Boskovic T, Stanic J, Pena-Karan S, et al. Pneumothorax after transthoracic needle biopsy of lung lesions under CT guidance. J Thorac Dis. 2014;6(Suppl 1):S99–107.

    PubMed  PubMed Central  Google Scholar 

  7. Ozturk K, Soylu E, Gokalp G, Topal U. Risk factors of pneumothorax and chest tube placement after computed tomography-guided core needle biopsy of lung lesions: a single-centre experience with 822 biopsies. Pol J Radiol. 2018;83:e407–14.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Laurent F, Michel P, Latrabe V, et al. Pneumothoraces and chest tube placement after CT-guided transthoracic lung biopsy using a coaxial technique: incidence and risk factors. AJR Am J Roentgenol. 1999;172:1049–53.

    Article  CAS  PubMed  Google Scholar 

  9. Topal U, Ediz B. Transthoracic needle biopsy: factors effecting risk of pneumothorax. Eur J Radiol. 2003;48:263–7.

    Article  PubMed  Google Scholar 

  10. Cox JE, Chiles C, McManus CM, Aquino SL, Choplin RH. Transthoracic needle aspiration biopsy: variables that affect risk of pneumothorax. Radiology. 1999;212:165–8.

    Article  CAS  PubMed  Google Scholar 

  11. Lee DS, Bak SH, Jeon YH, Kwon SO, Kim WJ. Perilesional emphysema as a predictor of risk of complications from computed tomography-guided transthoracic lung biopsy. Jpn J Radiol. 2019;37:808–16.

    Article  PubMed  Google Scholar 

  12. Incoronato M, Aiello M, Infante T, et al. Radiogenomic Analysis of Oncological Data: a technical Survey. Int J Mol Sci. 2017;18:805.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Just N. Improving tumour heterogeneity MRI assessment with histograms. Br J Cancer. 2014;9:111:2205–13.

    Article  Google Scholar 

  14. Wu G, Jochems A, Refaee T, et al. Structural and functional radiomics for lung cancer. Eur J Nucl Med Mol Imaging. 2021;48:3961–74.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Lambin P, Rios-Velazquez E, Leijenaar R, et al. Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer. 2012;48:441–6.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Baldwin D, O’Dowd E, Tietzova I, et al. Developing a Pan-european technical standard for a comprehensive high-quality lung cancer CT screening program. An ERS technical standard. Eur Respir J. 2023:61:2300128. https://doi.org/10.1183/13993003.00128.

    Article  PubMed  Google Scholar 

  17. Global initiative for chronic obstructive lung disease. The Global Strategy for Diagnosis, Management and Prevention of COPD (updated 2023). www.goldcopd.org. Accessed 1st June 2023.

  18. Graham BL, Steenbruggen I, Miller MR, et al. Standardization of Spirometry 2019 Update. An official American Thoracic Society and European Respiratory Society Technical Statement. Am J Respir Crit Care Med. 2019;200:e70–88.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Culver BH, Graham BL, Coates AL, et al. Recommendations for a standardized pulmonary function report. An Official American Thoracic Society Technical Statement. Am J Respir Crit Care Med. 2017;196:1463–72.

    Article  PubMed  Google Scholar 

  20. Davis MD, Walsh BK, Sittig SE, Restrepo RD. AARC clinical practice guideline: blood gas analysis and hemoximetry: 2013. Respir Care. 2013;58:1694–703.

    Article  PubMed  Google Scholar 

  21. de Margerie-Mellon C, de Bazelaire C, de Kerviler E. Image-guided biopsy in primary lung cancer: why, when and how. Diagn Interv Imaging. 2016;97:965–72.

    Article  PubMed  Google Scholar 

  22. Wang Z, Gu S, Leader JK, et al. Optimal threshold in CT quantification of emphysema. Eur Radiol. 2013;23:975–84.

    Article  PubMed  Google Scholar 

  23. Albano D, Messina C, Vitale J, Sconfienza LM. Imaging of Sarcopenia: old evidence and new insights. Eur Radiol. 2020;30(4):2199–208.

    Article  PubMed  Google Scholar 

  24. Szczypiński PM, Strzelecki M, Materka A, Klepaczko A. MaZda—a software package for image texture analysis. Comput Methods Prog Biomed. 2009;94:66–76.

    Article  Google Scholar 

  25. Strzelecki M, Szczypinski P, Materka A, Klepaczko A. A software tool for automatic classification and segmentation of 2D/3D medical images. Nucl Instrum Methods Phys Res A. 2013;702:137–40.

    Article  CAS  Google Scholar 

  26. Fruehwald-Pallamar J, Czerny C, Holzer-Fruehwald L, et al. Texture-based and diffusion-weighted discrimination of parotid gland lesions on MR images at 3.0 Tesla. NMR Biomed. 2013;26:1372–9.

    Article  PubMed  Google Scholar 

  27. Meyer HJ, Schob S, Höhn AK, Surov A. MRI texture analysis reflects histopathology parameters in thyroid Cancer - A First Preliminary Study. Transl Oncol. 2017;10:911–6.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Zhao Y, Bao D, Wu W, Tang W, Xing G, Zhao X. Development and validation of a prediction model of pneumothorax after CT-guided coaxial core needle lung biopsy. Quant Imaging Med Surg. 2022;12:5404–19.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Elshafee AS, Karch A, Ringe KI, et al. Complications of CT-guided lung biopsy with a non-coaxial semi-automated 18 gauge biopsy system: frequency, severity and risk factors. PLoS ONE. 2019;14:e0213990.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Chami HA, Faraj W, Yehia ZA, et al. Predictors of pneumothorax after CT-guided transthoracic needle lung biopsy: the role of quantitative CT. Clin Radiol. 2015;70:1382–7.

    Article  CAS  PubMed  Google Scholar 

  31. Lendeckel D, Kromrey ML, Ittermann T, et al. Pulmonary emphysema is a predictor of pneumothorax after CT-guided transthoracic pulmonary biopsies of pulmonary nodules. PLoS ONE. 2017;12:e0178078.

    Article  PubMed  PubMed Central  Google Scholar 

  32. An W, Zhang H, Wang B, Zhong F, Wang S, Liao M. Comparison of CT-Guided core needle biopsy in Pulmonary Ground-Glass and solid nodules based on propensity score matching analysis. Technol Cancer Res Treat. 2022;21:15330338221085357.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Zhou W, Yu X, Song Y, et al. Risk of pneumothorax caused by computerized tomography-guided percutaneous core needle biopsy of the lung in elderly and young patients. J Cancer Res Ther. 2021;17:1186–91.

    Article  PubMed  Google Scholar 

  34. Peng B, Deng Z, Wang Y, et al. The risk of immediate pneumothorax after CT-guided lung needle biopsy: pleural tail sign as a novel factor. Quant Imaging Med Surg. 2023;13:707–19.

    Article  PubMed  Google Scholar 

  35. Ruud EA, Stavem K, Geitung JT, et al. Predictors of pneumothorax and chest drainage after percutaneous CT-guided lung biopsy: a prospective study. Eur Radiol. 2021;31:4243–52.

    Article  PubMed  Google Scholar 

  36. Nagarajan G, Doshi P, Bardeskar NS, Kulkarni A, Punamiya A, Tongaonkar H. Association between Sarcopenia and postoperative complications in patients undergoing surgery for gastrointestinal or hepato-pancreatico-biliary cancer. J Surg Oncol. 2023;128:708–10

  37. Xiong B, Yang C, Zhu X et al. The added value of Sarcopenia on existing risk scores to Predict Mortality after TIPS Placement: a Multicenter Study. Acad Radiol 2023;S1076-6332(23)00123-X.

  38. Hasenauer A, Forster C, Hungerbühler J, et al. CT-Derived Sarcopenia and outcomes after Thoracoscopic Pulmonary Resection for Non-small Cell Lung Cancer. Cancers (Basel). 2023;15:790.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

AF was supported by the postdoctoral fellowship “MetaRot program” (clinician scientist program) from the Federal Ministry of Education and Research (BMBF), Germany (FKZ 01EO1501, IFB Adiposity Diseases), a research grant from the “Mitteldeutsche Gesellschaft für Pneumologie (MDGP) e.V.” (2018-MDGP-PA-002), a junior research grant from the Medical Faculty, University of Leipzig (934100-012), a graduate fellowship from the “Novartis Foundation”, and by the “PETictCAC” project (ERAPerMed_324), which was funded with tax funds based on the budget passed by the Saxon State Parliament (Germany) under the frame of ERA PerMed (Horizon 2020 Research and Innovation Framework Programme of the European Commission Research Directorate-General, Grant Agreement No. 779282).

Funding

Open Access funding enabled and organized by Projekt DEAL.

Author information

Author notes

  1. Jakob Leonhardi and Ulrike Dahms contributed equally to this work.

Authors and Affiliations

  1. Department of Diagnostic and Interventional Radiology, University of Leipzig, 49341/9717400, Leipzig, Germany

    Jakob Leonhardi, Ulrike Dahms, Benedikt Schnarkowski, Sebastian Ebel, Gordian Prasse, Timm Denecke & Hans-Jonas Meyer

  2. Department of Anesthesiology and Intensive Care Medicine, University Hospital Leipzig, Leipzig, Germany

    Manuel Florian Struck

  3. Department of Pathology, University Hospital Leipzig, University of Leipzig, Leipzig, Germany

    Anne-Kathrin Höhn

  4. Department of Thoracic Surgery, University Hospital Leipzig, University of Leipzig, Leipzig, Germany

    Sebastian Krämer

  5. Department of Respiratory Medicine, University Hospital Leipzig, University of Leipzig, Leipzig, Germany

    Armin Frille

Authors
  1. Jakob Leonhardi
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ulrike Dahms
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Benedikt Schnarkowski
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Manuel Florian Struck
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Anne-Kathrin Höhn
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Sebastian Krämer
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Sebastian Ebel
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Gordian Prasse
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Armin Frille
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Timm Denecke
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Hans-Jonas Meyer
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

HJM and UD: designed the study, planned the experimental work, and analyzed the data. JL, UD, AKH, AF, SK, BS, GP, and SE: performed experimental work and analyzed the data. HJM and UB: wrote the manuscript. TD, MFS and AF: critically reviewed the manuscript. All Authors approved the submission of this manuscript.

Corresponding author

Correspondence to Hans-Jonas Meyer.

Ethics declarations

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

The study was approved by the University of Leipzig Review Board.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Leonhardi, J., Dahms, U., Schnarkowski, B. et al. Impact of radiomics features, pulmonary emphysema score and muscle mass on the rate of pneumothorax and chest tube insertion in CT-guided lung biopsies. Respir Res 25, 320 (2024). https://doi.org/10.1186/s12931-024-02936-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12931-024-02936-6

Keywords

Respiratory Research

ISSN: 1465-993X

Contact us