Sage Journals: Discover world-class research

Abstract

Objective

To analyze independent risk factors for postoperative venous thromboembolism (VTE) in patients after thoracoscopic lobectomy and establish a risk model for predicting thrombosis.

Methods

We selected 135 patients who underwent thoracoscopic lobectomy and assigned them to a VTE group (n = 31) and a non-VTE group (n = 104) according to the presence or absence of postoperative VTE. Univariate and multivariate logistic regression models were built to identify independent risk factors for postoperative thrombosis. A nomogram to predict the risk of thrombosis was generated, with the predictive performance of the model being evaluated by plotting calibration curve and the receiver operating characteristic (ROC) curve.

Results

D-dimer on the first postoperative day and Caprini scores were independent high-risk factors for VTE after thoracoscopic lobectomy as revealed by univariate and multivariate logistic regression analyses (p < .05). The nomogram was evaluated, and the fitting degree of calibration curve indicated good accuracy of the model in predicting thrombosis. The area under the curve (AUC) value was 0.8654, suggesting favorable diagnostic efficacy of the model.

Conclusion

The D-dimer on the first postoperative day and Caprini scores were independent risk factors for VTE after thoracoscopic lobectomy. The nomogram that was generated to predict the risk of thrombosis had good accuracy and diagnostic efficacy, which can guide individual risk assessment and management decisions in clinical practice.

Keywords

thoracoscopic lobectomy thrombus high-risk factors nomogram

Introduction

Current treatments for lung cancer (LC) include surgical resection, immunotherapy, radiotherapy, and targeted therapy, among which pneumonectomy is the best option for patients with early-stage LC.¹ In the early 1990s, video-assisted thoracoscopic surgery (VATS) lobectomy was, for the first time, used for LC resection and became the gold standard for LC surgery as its safety and feasibility were confirmed by mounting clinical evidence.² Thoracoscopic lobectomy can relieve pain and accelerate recovery postoperatively in patients and is associated with lower complication rates, mortality, and costs into the bargain.^3–5 Patients undergoing thoracoscopic lobectomy have a complication rate of 58% lower and an overall survival rate higher than those undergoing open lobectomy (p < .001).^6,7 On this basis, thoracoscopic lobectomy is deemed safe and effective as a high-performance alternative to open lobectomy.

Thoracoscopic lobectomy is beneficial in many aspects, while postoperative complications may still occur to varying degrees. Venous thromboembolism (VTE) is a serious complication after thoracoscopic lobectomy,^8,9 and patients with thrombosis are at high risk of critical symptoms including pulmonary embolism, infarction, or chronic thromboembolic pulmonary hypertension. As a case report presented, a patient with LC who underwent right lower lobectomy and mediastinal lymphadenectomy developed thrombosis in the right lower lobar pulmonary artery stump three months later.¹⁰ A patient with small cell LC who underwent thoracoscopic left upper lobectomy experienced epigastric abdominal pain and vomiting 16 days later and then received a diagnosis of propagation of thrombus in the stump of the left superior pulmonary vein complicated with splenic infarction as shown by contrast-enhanced computed tomography (CT).¹¹ Additionally, Kitajima et al¹² reported a case of acute cerebral infarction due to stasis of blood flow in the left upper pulmonary vein stump around 30 h after the left upper lobectomy. Thrombosis is often implicated in poor outcomes. As demonstrated by Chen et al.,¹³ a lower survival rate was observed in patients with thrombosis than in those without thrombosis after pneumonectomy. In conclusion, thrombosis is prone to occur after thoracoscopic lobectomy and increases the risk of death in patients, thus high priority should be accorded in clinical practice. But studies on predicting the risk of thrombosis after thoracoscopic lobectomy are insufficient to fully elucidate risk factors.

In this study, the clinical data of patients who underwent thoracoscopic lobectomy at Dongyang People's Hospital from January 2017 to December 2022 were statistically analyzed to investigate the risk factors for postoperative VTE. A nomogram was generated in the combination of factors to predict the onset risk, so as to guide the stratification of high-risk patients for adopting effective prevention, diagnosis, and treatment, thereby accelerating postoperative recovery.

Methods

Subjects

Patients who underwent thoracoscopic lobectomy at Dongyang People's Hospital from January 2017 to December 2022 were selected. Information on clinical characteristics of patients was collected, including age, gender, blood type, body mass index (BMI), diabetes, hypertension, coronary heart disease, preoperative D-dimer, preoperative carcinoembryonic antigen (CEA), preoperative forced expiratory volume in 1 s (FEV1), Caprini score, operation time, surgical site, D-dimer on the first postoperative day, pathological type, and pathological stage. Patients who met the following inclusion criteria were deemed eligible: (1) patients were diagnosed as LC by pathological examination; (2) patients received thoracoscopic lobectomy; (3) patients received a diagnosis of postoperative thrombus by vascular ultrasound. Exclusion criteria included: (1) idiopathic thrombocytopenic purpura; (2) cerebral infarction or myocardial infarction in the past year; (3) perioperative use of anticoagulant drugs due to coagulopathy or other causes; (4) preoperative thrombosis; (5) incomplete clinical data. As per the inclusion and exclusion criteria, 135 patients who underwent thoracoscopic lobectomy were included and assigned to a VTE group (n = 31) and a non-VTE group (n = 104) according to the presence or absence of postoperative VTE.

Construction and Evaluation of Nomogram

The risk factors of thrombosis after thoracoscopic lobectomy obtained by multivariate logistic analysis were the basis of nomogram construction. Discrimination evaluation of the model was done by plotting receiver operating characteristics (ROC) curves and calculating the area under the curve (AUC). Regarding the calibration evaluation of the model, the calibration curve of the predicted complication probability by nomogram and the actual complication probability was plotted to validate the consistency of the model.

Statistics

Statistical analysis was performed on SPSS 24.0 software. Measurement data were expressed as mean ± standard deviation, and an independent sample t-test was used; enumeration data were expressed as N (%), and a chi-square test was used. Logistic regression model was established with risk factors at p < .1, and independent risk factors were determined. Univariate and multivariate logistic regression models were used for nomogram construction. The Caprini scores were calculated and summarized postoperatively based on the modified Caprini risk assessment model. p < .05 meant statistical significance.

Results

Patient Baseline Characteristics

135 patients who underwent thoracoscopic lobectomy were included and assigned to the VTE group and the non-VTE group according to the presence or absence of thrombosis. The baseline information of patients was listed in Table 1. Univariate analysis of patient clinical data presented significant differences in age, gender, coronary heart disease, surgical site, D-dimer on the first postoperative day, preoperative FEV1, and Caprini score between the two groups (p < .1). No significant differences were seen in other factors (p > .1).

Table 1.

Patient Baseline Characteristics.

Patients Characteristics	VTE group (n = 31)	Non-VTE group (n = 104)	p
Age	64.7 ± 6.0	60.1 ± 9.4	.002
Gender:			.061
Female	22 (71.0%)	54 (58.7%)
Male	9 (29.0%)	50 (41.3%)
Blood type:			.112
A	13 (41.9%)	36 (34.6%)
AB	5 (16.1%)	5 (4.8%)
B	6 (19.4%)	31 (29.8%)
O	7 (22.6%)	32 (30.8%)
Hypertension:			.250
No	19 (61.3%)	75 (72.1%)
Yes	12 (38.7%)	29 (27.9%)
Diabetes:			.270
No	27 (87.1%)	97 (93.3%)
Yes	4 (12.9%)	7 (6.7%)
Coronary heart disease:			.097
No	27 (87.1%)	87 (83.7%)
Yes	4 (12.9%)	17 (16.3%)
Preoperative D-dimer:	0.88 ± 0.51	0.91 ± 1.53	.642
D-dimer post first	2.83 ± 2.11	1.26 ± 0.77	.000
Operation time:	138.1 ± 46.6	149.9 ± 39.4	.163
Surgical site:			.086
left	7 (22.6%)	41 (39.4%)
right	24 (77.4%)	63 (60.6%)
Caprini Score:			.001
6	0 (0.0%)	1 (1.0%)
7	4 (12.9%)	54 (51.9%)
8	23 (74.2%)	42 (40.4%)
9	4(12.9%)	7(6.7%)
Pathological type:			.207
adenocarcinoma	30 (96.8%)	93 (89.4%)
else	1 (3.2%)	11 (10.6%)
Pathological stage:			.994
I	28 (90.3%)	94 (90.4%)
II	2 (6.5%)	7 (6.7%)
III	1 (3.2%)	3 (2.9%)
BMI	23.5 ± 2.8	22.8 ± 2.6	.200
Preoperative FEV1	2.12 ± 0.47	2.36 ± 0.61	.049
Preoperative CEA	3.72 ± 2.27	3.41 ± 3.93	.674

*VTE: venous thromboembolism; BMI: body mass index; FEV1: forced expiratory volume in 1 s; CEA: carcinoembryonic antigen.

Note: The bold value indicates p < .05, which is statistically significant.

Results of Logistic Regression Analysis

By using variables with statistical differences between the two groups as independent variables and thrombosis as the dependent variable, multivariate logistic regression analysis was performed, and a logistic model was built:

\begin{aligned} l o g i t (P) = & - 12.269 - 0.063 x_{A g e} - 0.732 x_{G e n d e r} \\ - 0.160 x_{C o r o n a r y h e a r t d i s e a s e} + 1.096 x_{D - 2 - m e r p o s t 1 s t} \\ + 0.728 x_{S u r g i r a l s i t e} \\ + .697 x_{C a p r i n S c o r e} - 0.173 x_{P r e o p e r a t i v e F E V 1} \end{aligned}

D-dimer on the first postoperative day (p = .000) and Caprini score (p = .010) were independent risk factors for the development of VTE after thoracoscopic lobectomy (Table 2).

Table 2.

Variables in the Logistic Regression Equation.

Factors	β	SE	p	OR	95%CI
Age	−0.063	0.052	.227	0.939	0.849–1.040
Gender	−0.732	0.667	.272	0.480	0.130–1.776
Coronary heart	−0.160	0.703	.820	0.852	0.215–3.383
D-dimer post first	1.096	0.272	.000	2.991	1.755–5.097
Surgical site	0.728	0.588	.216	2.071	0.654–6.558
Caprini Score	1.697	0.655	.010	5.457	1.513–19.689
Preoperative FEV1	−0.173	0.710	.059	0.808	0.209–3.383
Constant	−12.269	4.429	.006	0.000

Note: The bold value indicates p < .05, which is statistically significant.

Establishment of a Nomogram for the Risk Model

A nomogram was developed to predict the risk of VTE after thoracoscopic lobectomy based on the logistic regression model (Figure 1). As depicted by the nomogram, the development of postoperative thrombosis was mostly affected by preoperative D-dimer, followed by Caprini score. The preoperative D-dimer and Caprini scores as single scores were calculated, and the total score was accumulated. The probability of postoperative thrombosis could be predicted according to the total score. The calibration curve of the nomogram was plotted and showed a tendency to be a standard curve, indicating that the predicted value was basically in line with the actual situation (Figure 2). As analyzed on the ROC curve, the AUC value was 0.8654, suggesting that the model had favorable accuracy and discrimination (Figure 3).

Figure 1.

Nomogram of high-risk factors for thrombosis after thoracoscopic lobectomy.

Figure 2.

Calibration curve of the nomogram.

Figure 3.

ROC curve of the nomogram.

Discussion

Thoracoscopic lobectomy is the main approach for LC treatment and can achieve satisfactory clinical efficacy, but postoperative complications such as VTE are prone to arise. Risk factors for VTE include surgical approach, anesthesia, and patient stress.^14,15 Clinical interests have grown up around the theme of causes of postoperative thrombosis after thoracoscopic lobectomy, but no consensus is reached on the independent risk factors for postoperative thrombosis after thoracoscopic lobectomy and its prevention. In this study, D-dimer on the first postoperative day and Caprini scores were independent risk factors for postoperative VTE after thoracoscopic lobectomy as identified by univariate and multivariate analyses.

D-dimer is associated with the prognoses of LC patients, and elevated postoperative serum D-dimer levels increase the risk of postoperative thrombosis in LC patients. Wang et al.¹⁶ reported that up to 87.17% (523/600) of patients undergoing lung tumor surgery have postoperative VTE, and postoperative D-dimer levels differ substantially between the VTE group and the non-VTE group. Additionally, Yao et al.¹⁷ confirmed that at 24 h after thoracoscopic surgery, hypercoagulation, and progressively elevated D-dimer are observed, but the impact is less than thoracotomy. D-dimer is the smallest fibrinolysis-specific degradation product known in the circulation,¹⁸ and its elevated level is generally indicative of hypercoagulation and secondary hyperfibrinolysis.¹⁹ Serial postoperative testing of D-dimer levels combined with risk factors associated with thrombosis allows well evaluation of VTE occurrence. Caprini score is a valid and reliable predictive tool for assessing the risk of thrombosis. Zhou et al.²⁰ revealed that the Caprini score identifies 84.3% of VTE cases, whereas the Padua Prediction score only identifies 49.1% of VTE cases, besides, the Caprini risk level is associated with in-hospital and 6-month mortality. Chen et al.²¹ observed VTE in 10 of 117 patients who underwent pneumonectomy, and Caprini score is proportional to the VTE ratio, indicating the superior performance of Caprini score in predicting VTE. Thence, serial assessment of perioperative Caprini scores is effective in reducing the incidence of thrombosis and mortality of critically ill patients, improving patient prognosis and quality of life.²² Together these findings confirm that D-dimer levels on the first postoperative day and Caprini scores can be key indicators for postoperative thrombotic screening, suggesting that a combination of these two indicators is highly sensitive for identifying patients who may benefit from prevention.

In this study, a predictive model was successfully constructed with D-dimer levels on the first postoperative day and Caprini scores on the basis of multivariate regression analysis. Meanwhile, the calibration curve and ROC curve suggested that the nomogram had favorable predictive performance and discrimination. Clinicians can stratify patients at high risk of postoperative thrombosis referring to this model and be prepared to take targeted preventive measures. High-risk patients are encouraged to receive regular checkups for early detection and treatment. A growing body of research is done on the prevention of postoperative thrombosis after thoracoscopic lobectomy. Chen et al.²³ determined that self-heating calf sleeves (SHCSs) prevent postoperative VTE in patients undergoing thoracoscopic lobectomy, and SHCS is a satisfactory prophylaxis method to improve venous blood flow and thus alleviate venous stasis in the lower extremities. Nakano et al.²⁴ reported that cases with thrombus formation in the pulmonary vein stump as shortened by ligating the pulmonary vein at the pericardial reflection after left upper lobectomy are substantially fewer than those not treated with the ligation method (p = .010), demonstrating that the proximal ligation method is markedly linked to less thrombus formation in the pulmonary vein stump. Thus, stratifying patients at high risk for postoperative thrombosis after thoracoscopic lobectomy facilitates the implementation of prevention measures and reduces patient harm and economic burden.

Taken together, D-dimer level on the first postoperative day and Caprini score are independent risk factors for VTE after thoracoscopic lobectomy. In this study, the predictive model shows good accuracy and discrimination and is beneficial for clinical screening of high-risk groups, thus interventions can be taken correspondingly to improve patient prognosis. This model is valuable in clinical applications. However, this study still has some limitations. This study is a single center, retrospective study, which may have bias. Secondly, the limited sample size and the lack of external validation of the nomogram allow unavoidable errors. A multi-center, large sample-size clinical study is ongoing to optimize and validate the model.

Footnotes

Authors’ Contributions

(I) Conception and design: Junkai Qiao. (II) Administrative support: Yueming Wu. (III) Provision of study materials or patients: Yueming Wu. (IV) Collection and assembly of data: Jiang Feng. (V) Data analysis and interpretation: Weiwei Hu. (VI) Manuscript writing: Junkai Qiao. (VII) Final approval of manuscript: All authors.

Availability of Data and Materials

The data and materials in the current study are available from the corresponding author on reasonable request.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethics Approval

Ethical approval to report this case was obtained from Dongyang People’s Hospital (2020-YX-077).

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Dongyang Municipal Science and Technology Program (grant number 20-323).

Informed Consent

Written informed consent was obtained from a legally authorized representative for anonymized patient information to be published in this article.

ORCID iD

Yueming Wu

Correction (May 2023):

The funding and Introduction text has been edited slightly since its original publication.

References

Boujibar

Gillibert

Gravier

, et al. Performance at stair-climbing test is associated with postoperative complications after lung resection: A systematic review and meta-analysis. Thorax 2020;75(9):791–797.

Sihoe

ADL

. Video-assisted thoracoscopic surgery as the gold standard for lung cancer surgery. Respirology 2020;25(Suppl 2):49–60.

Bendixen

Jorgensen

Kronborg

Andersen

Licht

. Postoperative pain and quality of life after lobectomy via video-assisted thoracoscopic surgery or anterolateral thoracotomy for early stage lung cancer: A randomised controlled trial. Lancet Oncol 2016;17(6):836–844.

Swanson

Meyers

Gunnarsson

, et al. Video-assisted thoracoscopic lobectomy is less costly and morbid than open lobectomy: A retrospective multiinstitutional database analysis. Ann Thorac Surg 2012;93(4):1027–1032.

Agostini

Lugg

Adams

, et al. Postoperative pulmonary complications and rehabilitation requirements following lobectomy: A propensity score matched study of patients undergoing video-assisted thoracoscopic surgery versus thoracotomydagger. Interact Cardiovasc Thorac Surg 2017;24(6):931–937.

Napolitano

Sparks

Werba

Rosenfeld

Antevil

Trachiotis

. Video-assisted thoracoscopic surgery lung resection in United States veterans: Trends and outcomes versus thoracotomy. Thorac Cardiovasc Surg 2022;70(4):346–354.

Merritt

Abdel-Rasoul

D’Souza

Kneuertz

. Lymph node upstaging for robotic, thoracoscopic, and open lobectomy for stage T2–3N0 lung cancer. Ann Thorac Surg 2022;115(1):175–182.

Morinaga

Nii

Sakamoto

Inoue

Mitsutake

Hanada

. Revascularization for in-hospital acute ischemic stroke after video-assisted thoracic surgery: Report of 2 cases and literature review. World Neurosurg 2019;129:28–33.

Fujii

Mori

Kambara

, et al. Pulmonary vein thrombosis and cerebral infarction after video-assisted thoracic surgery of the left upper lobe: A case series. JA Clin Rep 2020;6(1):71.

10.

Wieteska-Milek

Winiarczyk

Kupis

. Treatment of pulmonary artery stump thrombosis after lobectomy: A case report and literature review. Adv Respir Med 2021;89(3):311–315.

11.

Harada

Yunoki

Sugiyama

, et al. Left superior pulmonary venous thrombosis complicated with splenic infarction after video-assisted thoracoscopic left upper lobectomy. J Cardiol Cases 2017;16(1):1–4.

12.

Kitajima

Otsuka

Lefor

Sanui

. Acute cerebral infarction in a patient with an epidural catheter after left upper lobectomy: A case report. BMC Anesthesiol 2019;19(1):27.

13.

Chen

Zhang

Yang

Zhai

Wang

. Prognostic significance of arterial and venous thrombosis in resected specimens for non-small cell lung cancer. Thromb Res 2015;136(2):451–455.

14.

Qian

Dai

. Analysis of influencing factors of community-acquired pneumonia with deep venous thrombosis in elderly communities. Ann Palliat Med 2022;11(6):1997–2006.

15.

Liu

Tang

Tan

. Risk factors associated with catheter-related venous thrombosis: A meta-analysis. Public Health 2022;205:45–54.

16.

Wang

Zhao

, et al. Risk factors and clinical significance of D-dimer in the development of postoperative venous thrombosis in patients with lung tumor. Cancer Manag Res 2020;12:5169–5179.

17.

Yao

Shen

Zhou

Yang

Huang

. Efficacy of thoracoscopic surgery in the treatment of lung cancer in the perioperative period and its effects on serum D-dimer. Oncol Lett 2018;15(4):4397–4403.

18.

Olson

. D-dimer: An overview of hemostasis and fibrinolysis, assays, and clinical applications. Adv Clin Chem 2015;69:1–46.

19.

Chai

, et al. Expression of RSK4, CD44 and MMP-9 is upregulated and positively correlated in metastatic ccRCC. Diagn Pathol 2020;15(1):28.

20.

Zhou

, et al. Assessment of the risk of venous thromboembolism in medical inpatients using the Padua prediction score and caprini risk assessment model. J Atheroscler Thromb 2018;25(11):1091–1104.

21.

Chen

Cai

Zhang

, et al. Prevalence of venous thromboembolism and evaluation of a modified caprini risk assessment model: A single-centre, prospective cohort study involving patients undergoing lung resections for bronchiectasis. Thromb J 2022;20(1):43.

22.

Cronin

Dengler

Krauss

, et al. Completion of the updated caprini risk assessment model (2013 version). Clin Appl Thromb Hemost 2019;25:1076029619838052.

23.

Chen

Song

Wang

. Efficacy of self-heating calf sleeves for preventing deep vein thrombosis in lung cancer patients who undergo video-assisted thoracoscopic surgery lobectomy. Ann Palliat Med 2020;9(5):2693–2698.

24.

Nakano

Kaneda

Kawaura

Kitawaki

Murakawa

. Ligating the pulmonary vein at the pericardial reflection is useful for preventing thrombus formation in the pulmonary vein stump after left upper lobectomy. Gen Thorac Cardiovasc Surg 2019;67(5):450–456.

Risk Factor Analysis of Postoperative Venous Thromboembolism in Patients After Thoracoscopic Lobectomy

Abstract

Objective

Methods

Results

Conclusion

Keywords

Introduction

Methods

Subjects

Construction and Evaluation of Nomogram

Statistics

Results

Patient Baseline Characteristics

Results of Logistic Regression Analysis

Establishment of a Nomogram for the Risk Model

Discussion

Footnotes

Authors’ Contributions

Availability of Data and Materials

Declaration of Conflicting Interests

Ethics Approval

Funding

Informed Consent

ORCID iD

Correction (May 2023):

References