Abstract
Background:
Venous thromboembolism (VTE) is a well-recognized complication following gastrointestinal cancer surgery, particularly early postoperatively. The incidence and risk factors of VTE within 1-year after esophageal (including esophago-gastric junction) (ECS) and gastric (GCS) cancer surgeries, and especially its impact on 1-year global mortality, are yet under-explored.
Methods:
This nationwide observational population-based cohort study used data extracted from all patients undergoing ECS and GCS in France between 1 January 2015 and 31 December 2017. Multivariate logistic regression was used to identify risk factors for 90 postoperative days (POD) VTE (OR 95% CI). Cox proportional hazards models investigated the impact of 1-year postoperative VTE on 1-year global mortality [HR (95% CI)].
Results:
During the study period, 8005 patients underwent ECS (N=3429) or GCS (N=4576) (31.8% female; 66.7±12.1 years old). Majority (N=4951) of patients had preoperative treatment (chemotherapy or radiochemotherapy). Ninety POD incidence of VTE were 4.7% (ECS=6.2%) (GCS=3.6%) (44.7% during first hospitalization, 19.0% needing readmission, and 36.3% ambulatory management). Main risk factors were three and two field esophagectomy [3.6 (2.20–5.83) and 2.2 (1.68–3.0)], obesity [1.9 (1.40–2.58)] and history of VTE [5.1 (2.72–9.45)]. Late-onset VTE rates (occurring between the 6th and 12th month) represented 1.80 and 1.46% of the overall ECS and GCS groups. Patients with VTE within 1-year had higher risks of 1-year global mortality: (2.04 1.52; 2.73) and 2.71 (2.09; 3.51), respectively.
Conclusion:
Our extensive analysis of a nationwide database highlights the significant risk of postoperative VTE after ECS and GCS, persisting within 90 POD and up to 1-year. Crucially, a higher risk of global mortality within 1-year for patients experiencing early or late VTE was found. These findings could advocate for further research into extended prophylactic regimens, particularly for those most at risk.
Keywords: cancer, esophageal surgery, gastric surgery, global mortality, postoperative courses, pulmonary embolism, venous thromboembolism disease
Introduction
Highlights
Nationwide analysis of the risk factors associated with venous thromboembolism post upper gastrointestinal tract surgery for cancer and its association with 1-year global mortality.
Comprehensive use of the French SNDS database to obtain robust and extensive data within one postoperative years.
Identification of significant factors during preoperative evaluations and postoperative care in patients undergoing such at-risk surgery.
Cancer-related thrombosis remains one of the major causes of cancer mortality. Its mechanisms are multiple: direct invasion of blood vessels by the tumor, release of tumor-specific prothrombotic mediators into the circulation, activation of the endothelium (chemotherapy, surgery, and radiotherapy)1.
Among the major types of cancer, digestive cancers are the most common worldwide, and each has a different risk profile for venous thromboembolism (VTE), with cancers of the upper digestive tract recognized as being particularly at risk2,3. Surgery is often the preferred treatment for localized to locally advanced forms, at the cost of complex surgery that often requires extensive reconstruction. This is particularly the case for oncological surgery of the esophagus (including oesophago-gastric junction) (ECS) and stomach (GCS)3.
In addition, modern gastrointestinal oncology surgery of the esophagus and stomach is often preceded and/or followed by chemotherapy or radiochemotherapy to improve overall survival4,5. However, these factors are in themselves independent additional risk factors for VTE, leading to a theoretical excess risk of VTE preoperatively, as well as early and late postoperatively1,6,7.
Numerous large retrospective studies have investigated the risk of VTE after gastrointestinal surgery, but none have focused on this risk in the year following ECS or GCS3,8.
The objective of our study was to investigate the risk factors for VTE within 90 days following ECS and GCS, and to describe and assess the role of 1-year postoperative VTE as an independent risk factor for global mortality within the same timeframe. This study was conducted in a French national cohort of patients who underwent surgery between 1 January 2015 and 31 December 2017, utilizing data from the national health data system.
Patients and methods
The French national health insurance system database (SNDS)
The SNDS covers the entire French population (>66.6 million inhabitants in 2015), including low-income patients, and is considered to be one of the largest healthcare databases in the world. As a medico-administrative database, it is used to refund outpatient medical expenditures to the French population, as well as the cost of hospital stays in public and private hospitals. Briefly, the SNDS contains comprehensive data on all reimbursements for health-related expenditures, as well as demographic data and medical information on all admissions to French public and private hospitals (Diagnosis-related group model, according to the International Statistical Classification of Diseases, 10th revision–ICD-10) starting from 2005. Analyses of the SNDS database for the present study received the ethical and methodological approval of the French Health Data Hub (Entrepot des données de santé ) and the French Expert Committee for Research, Studies and Evaluations in the Field of Health (Comité d’Expertise pour les Recherches, les Études et les Évaluations dans le domaine de la Santé, N° T32257572020122). Finally, this study was authorized by the French personal data protection authority (Commission Nationale Informatique et Libertés, CNIL, N° 921140). No informed patient consent was required because this database is anonymous and all analyses were performed on the devoted secure portal of the French National Health Insurance. The SNDS is often used for scientific publications, as this comprehensive database allows systematic follow-up. Part of our team has previously published in the field of digestive surgery using data from the SNDS9–11.
Population
All patients undergoing ECS and GCS with a primary diagnosis of esophageal (EGJ included) or gastric cancer in France between 1 January 2015 and 31 December 2017, were identified (eTable 1, Supplemental Digital Content 2, http://links.lww.com/JS9/B546).
Patients with any of the following criteria were not included in our data extraction: age less than 18 years, palliative surgery for cancer, primary diagnosis of neoplasms of uncertain or unknown behavior (ICD-10 codes D37 to D48), and digestive surgery for secondary malignant neoplasm. Patients with an anticoagulant reimbursement (LWMH and Oral Direct Anticoagulant) in the 3 months prior to surgery were also excluded.
This cohort study follows strengthening the reporting of cohort, cross-sectional and case–control studies in surgery (STROCSS) (Supplemental Digital Content 1, http://links.lww.com/JS9/B545) criteria12.
Data collection and definitions
Data were extracted from 1 January 2010 to 31 December 2018, to have data within the 5 years before surgery and 1-year after surgery for each patient. They included baseline characteristics: age, sex, vital status, obesity status (BMI ≥30 kg/m²), comorbidities (eTable 2, Supplemental Digital Content 2, http://links.lww.com/JS9/B546) (including the Charlson score13 ), possible preoperative treatment within 90 days before surgery (chemotherapy, radiochemotherapy/radiotherapy) (eTable 3, Supplemental Digital Content 2, http://links.lww.com/JS9/B546) and history of VTE needing hospital admission within 5 years preoperatively (eTable 2, Supplemental Digital Content 2, http://links.lww.com/JS9/B546). Precarity was assessed using a validated geographic socioeconomic deprivation index and the presence of universal health insurance coverage for low-income families, as already reported. Type of surgical procedure was found using French nomenclature of surgical procedure (CCAM; Classification Commune des Actes Médicaux) (eTable4, Supplemental Digital Content 2, http://links.lww.com/JS9/B546). Surgical complications within first hospital stay were also recorded (eTable5, Supplemental Digital Content 2, http://links.lww.com/JS9/B546).
VTE definitions and assessment
Ninety-day VTE was defined as in-of-hospital and out-of-hospital mortality within 90 postoperative days. Occurrence of inpatient VTE [deep venous thrombosis or pulmonary embolism (PE)] was defined as the presence of VTE ICD-10 diagnosis codes (eTable 6, Supplemental Digital Content 2, http://links.lww.com/JS9/B546) during the hospitalization for ECS and GCS or any subsequent hospitalization.
Outpatient diagnoses was found using a combination of outpatient reimbursement for chest CT-scan and/or pulmonary scintigraphy and/or Doppler ultrasound of a lower limb plus at least one outpatient reimbursement for anticoagulant therapy during the first 30 days after imaging (eTable 7, Supplemental Digital Content 2, http://links.lww.com/JS9/B546).
Two time-frames were analyzed: the first 90 POD and the year following surgery after exclusion of the first 90 POD, to take into account a potential longer term risk of VTE.
Statistical analysis
Data management and statistical analysis were carried out using SAS 9.3 software (SAS Institute Inc.). The significance threshold was set to P=0.05.
A description of the samples for the covariates was provided for ECS and GCS. Comparisons between groups were performed using the χ2test for categorical variables and analysis of variance for continuous variables.
Several logistic regression models were tested to determine risk factors for 90 POD VTE in patients. Multivariate analysis was presented using Forrest plots. The Kaplan–Meier method was used to assess the survival without VTE after index surgery within 1-year.
The study employed a Cox proportional hazards model to investigate the impact of 1-year postoperative VTE on 1-year global mortality in patients undergoing surgery for ECS and GCS. The variable ‘VTE’ within the year following surgery was modeled as a time-dependent variable. The analysis was conducted in two stages: an unadjusted model and an adjusted model. Adjusted Cox model incorporated various covariates to control for potential confounding factors. These included the type of surgery performed, patient demographics (age and sex), Charlson comorbidity score, preoperative nutritional status, any history of VTE within 5 years before surgery, intraoperative complications.
Statistical significance was determined at the P<0.05 level, and hazard ratios (HR) with 95% CI were calculated to quantify the strength and precision of the association between postoperative VTE and 1-year mortality.
Data management and statistical analysis were performed by the Qualees company.
Results
Baseline characteristics
Between 1 January 2015 and 31 December 2017, 8005 patients (68.21% male, 66.7±12.1 years old) underwent ECS (N=3429) or GCS (N=4576). Rate of patients with Charlson score ≥5 was higher (P<0.001) for patients undergoing ECS (14.5%) than those undergoing GCS (10.8%). Overall, a majority of patients had preoperative (including perioperative scheme) chemotherapy (52.0%). The rate of preoperative radiochemotherapy was 22.3% for ECS. In ECS and GCS groups, majority of patients underwent 2-field esophagectomy (69.1%) and partial gastrectomy (49.6%), respectively. Baseline characteristics of the patients, details of the surgery performed and details of the different regimens of pre/post or perioperative treatment are shown in Table 1.
Table 1.
Patient and hospital characteristics stratified by cancer surgery in France between 2015 and 2017 (N=8005) (The SNDS data).
| Characteristics | Overall (N=8005) | Esophageal and EGJ cancer (N=3429) | Gastric cancer (N=4576) |
|---|---|---|---|
| Age (years) | 66.7±12.1 | 64.0±10.2 | 68.8±13.0 |
| Female | 2549 (31.8) | 651 (19.0) | 1898 (41.5) |
| Obesity, BMI unspecified | 671 (8.4) | 322 (9.4) | 349 (7.6) |
| Universal health insurance coverage for patients with low-income | 880 (11.0) | 363 (10.6) | 517 (11.3) |
| Geographical socioeconomic deprivation index | |||
| 1st quintile (less deprived) | 1285 (16.0) | 555 (16.2) | 730 (16.0) |
| 5th quintile (most deprived) | 1649 (20.6) | 701 (20.4) | 948 (20.7) |
| Annual volume of hospital | 18.0 [7.3–42.0] | 32.3 [13.7–55.3] | 11.8 [4.7–28.3] |
| Procedure performed in a low operative volume hospital (<5 cases/year) | 1361 (17.0) | 191 (5.6) | 1170 (25.6) |
| Charlson score (Categories) | |||
| 0–2 | 5352 (66.9) | 2132 (62.2) | 3220 (70.4) |
| 3–4 | 1661 (20.7) | 801 (23.4) | 860 (18.8) |
| ≥5 | 992 (12.4) | 496 (14.5) | 496 (10.8) |
| Malnutrition | 888 (11.1) | 414 (12.1) | 474 (10.4) |
| History of VTE | 72 (0.90) | 22 (0.64) | 50 (1.1) |
| Associated chemotherapya | |||
| None | 2945 (36.8) | 950 (27.7) | 1995 (43.6) |
| Preoperative | 1736 (21.7) | 1245 (36.3) | 491 (10.7) |
| Postoperative | 902 (11.3) | 186 (5.4) | 716 (15.6) |
| Perioperative | 2422 (30.3) | 1048 (30.6) | 1374 (30.0) |
| Associated radiochemotherapyb | |||
| None | 6968 (87.1) | 2582 (75.3) | 4386 (95.8) |
| Preoperative | 785 (9.8) | 766 (22.3) | 19 (0.42) |
| Postoperative | 244 (3.1) | 73 (2.1) | 171 (3.7) |
| Perioperative | 8 (0.10) | 8 (0.23) | 0 |
| Type of procedures | |||
| 2-field esophagectomy | 2403 (30.0) | 2370 (69.1) | 33 (0.72) |
| 3-field esophagectomy | 261 (3.3) | 260 (7.6) | 1 (0.02) |
| Transhiatal esophagectomy | 171 (2.1) | 166 (4.9) | 5 (0.11) |
| Total gastrectomy | 2828 (35.3) | 561 (16.4) | 2267 (49.5) |
| Partial gastrectomy | 2342 (29.3) | 72 (2.1) | 2270 (49.6) |
| Laparoscopic approach | 1904 (23.8) | 1275 (37.2) | 629 (13.8) |
| Length of stay (Days) | 16.0 [12.0–24.0] | 18.0 [14.0–28.0] | 15 [11.0–22.0] |
Data are presented as the mean (±SD) or median [IQR] for continuous variables, or numbers (%) for categorical variables.
Between 90 days before and/or after surgery.
Radiochemotherapy excluded.
EGJ, esophago-gastric Junction; IQR, interquartile range; SNDS, Système National des Données de Santé; VTE, venous thrombo-embolism.
Ninety-day postoperative VTE
For the entire cohort, the 90-day and 1-year mortality rates were 6.3 and 16.8%, respectively. The 90 POD VTE rate was 6.2 (ECS) and 3.6% (GCS), with a relative majority of diagnoses occurred during the initial hospitalization (45.3 and 44.0%). In 39.6% and 31.9% of cases, VTE management was conducted on an outpatient basis without the need for patient readmission (Table 2).
Table 2.
Postoperative VTE and others complications stratified by upper-GI cancer in France between 2015 and 2017 within first postoperative year (N=8005) (The SNDS data)a.
| Characteristics | Overall (N=8005) | Esophageal and EGJ cancer (N=3429) | Gastric cancer (N=4576) |
|---|---|---|---|
| Early events (≤90 POD) | |||
| Death | 505 (6.3) | 219 (6.4) | 286 (6.3) |
| Surgical complications | |||
| Infectious | 1136 (14.2) | 515 (15.0) | 621 (13.6) |
| Obstruction | 357 (4.5) | 97 (2.8) | 260 (5.7) |
| Haemorrhageb | 350 (4.4) | 93 (2.7) | 257 (5.6) |
| VTE | 378 (4.7) | 212 (6.2) | 166 (3.6) |
| During surgical stay | 169 (44.7) | 96 (45.3) | 73 (44.0) |
| Needing readmission | 72 (19.0) | 32 (15.1) | 40 (24.1) |
| Ambulatory management | 137 (36.3) | 84 (39.6) | 53 (31.9) |
| One year eventsc | |||
| Death | 1281 (16.8) | 539 (16.8) | 742 (16.8) |
| VTE | 501 (6.6) | 248 (7.7) | 253 (5.7) |
| Needing readmission | 229 (45.7) | 97 (39.1) | 132 (52.2) |
| Ambulatory management | 272 (54.3) | 151 (60.9) | 121 (47.8) |
| Late eventd | 129 (25.8) | 62 (25.0) | 67 (26.5) |
Data are presented as numbers (%).
Including haemothorax.
After exlusion of patients undergoing VTE during 90 POD and with prolonged primary surgical stay (>60 POD) (N=7631; N=3206; and N=4425, respectively).
Occurring between 180 and 365 POD.
EGJ, esophago-gastric Junction; POD, postoperative days; SNDS, Système National des Données de Santé; VTE, venous thromboembolism.
In univariate analysis, the most significant risk factors for the occurrence of 90 POD VTE were: ECS [1.75 (1.43 ; 2.17)], 3 and 2-field esophagectomy [3.10 (1.92; 5.01); 2.03 (1.52; 2.69)], presence of obesity [1.85 (1.37; 2.50)], a history of VTE [4.57 (2.48; 8.41)], and the occurrence of a serious surgical complication [1.60 (1.26; 2.03)]. The presence of preoperative treatment (radiochemotherapy or chemotherapy alone) was not identified as a significant risk factor (Table 3).
Table 3.
Univariate analysis of factors associated with 90 POD VTE after upper-GI cancer surgery performed in France between 2015 and 2017 (N=8005) (The SNDS data).
| Variable | Odds-ratio | (95% CI) | P |
|---|---|---|---|
| Age (Step=1-year increase) | 0.99 | [ 0.99–1.00 ] | 0.16 |
| Sex (Ref: male) | |||
| Female | 0.84 | [ 0.67–1.05 ] | 0.13 |
| Type of cancer (Ref: Esophageal and EGJ cancer) | |||
| Gastric cancer | 0.57 | [ 0.46–0.70 ] | <.001 |
| Surgical approach (Ref: Laparoscopy) | |||
| Laparotomy | 0.97 | [ 0.76–1.23 ] | 0.80 |
| Type of procedure (Ref: Partial gastrectomy) | |||
| 2-Field esophagectomy | 2.03 | [ 1.52–2.69 ] | <.001 |
| 3-Field esophagectomy | 3.10 | [ 1.92–5.01 ] | |
| Transhiatal esophagectomy | 1.11 | [ 0.48–2.60 ] | |
| Total gastrectomy | 1.42 | [ 1.06–1.90 ] | |
| Universal health insurance coverage for patients with low-income (Yes) | 0.77 | [ 0.53–1.10 ] | 0.14 |
| Geographical socioeconomic deprivation index (Ref: 1st quintile (less deprived)) | |||
| 5th quintile (most deprived) | 0.96 | [ 0.68–1.36 ] | 0.62 |
| Preoperative treatment (Ref: None) | |||
| Chemotherapy | 0.93 | [ 0.74–1.16 ] | 0.51 |
| Radiochemotherapy | 1.11 | [ 0.81–1.52 ] | |
| Charlson’s score (Ref: 0–2) | |||
| 3-4 | 0.96 | [ 0.74–1.25 ] | 0.95 |
| ≥5 | 0.99 | [ 0.72–1.37 ] | |
| Obesity (Yes) | 1.85 | [ 1.37–2.50 ] | <.001 |
| Malnutrition (Yes) | 0.86 | [ 0.61–1.22 ] | 0.40 |
| Private hospital (Yes) | 1.47 | [ 1.19–1.81 ] | <.001 |
| History of VTE (Yes) | 4.57 | [ 2.48–8.41 ] | <.001 |
| Surgical complication during hospital stay (Yes) | 1.60 | [ 1.26–2.03 ] | <.001 |
| Annual volume of caseloads per hospital (Ref: 0–19) | |||
| 20-39 | 1.20 | [ 0.93–1.56 ] | 0.18 |
| 40-59 | 0.92 | [ 0.66–1.29 ] | |
| ≥60 | 0.81 | [ 0.59–1.12 ] | |
EGJ, esophago-gastric junction; POD, postoperative days; SNDS, Système National des Données de Santé; VTE, venous thrombo-embolism.
In multivariate analysis, the most significant risk factors for the occurrence of 90 POD VTE were: 3 and 2-field esophagectomy [3.60 (2.20; 5.83); 2.2 (1.68; 3.00)], presence of obesity [1.9 (1.40;2.58)], a history of VTE [5.1 (2.72; 9.45)], and the occurrence of a serious surgical complication [1.7 (1.30; 2.11)] (Fig. 1).
Figure 1.
Forest plots showing the adjusted odds-ratio (OR) of 90 POD VTE in patients undergoing ECS and GCS. VTE, venous thromboembolism.
Description and assessment of 1-year postoperative VTE as an independent risk factor for global mortality within 1-year
In this section, VTE diagnoses performed before the 90 POD were excluded. The VTE rates for the entire cohort were 6.6, 7.7 (ECS), and 5.7% (GCS). In the ECS group, the majority of VTE cases were managed on an outpatient basis (60.9%). In the GCS group, most patients with VTE required hospitalization (52.2%). Late-onset VTE rates (occurring between the 6th and 12th month) were 25.0 and 26.5%, respectively. This represented 1.80 and 1.46% of the overall ECS and GCS groups (Table 2). Figure 2 depicts (Kaplan–Meier) VTE-free survival over the postoperative year for the ECS and GCS groups.
Figure 2.
One year postoperative survival probabilities without VTE for patients undergoing ECS or GCS. Patients dying from another cause were censored. ECS, esophageal (including oesophagogatsric junction) cancer surgery; GCS, gastric cancer surgery; VTE, venous thromboembolism.
In the unadjusted and adjusted models, patients undergoing GCS with 1-year postoperative VTE were at higher risk (HR) of 1-year global mortality times compared to those without VTE: 2.92 [95% CI (2.26–3.77), P<.001], 2.71 [95% CI (2.09–3.51), P<.001], respectively.
In the unadjusted and adjusted models, patients undergoing ECS with 1-year postoperative VTE were at higher risk (HR) of 1-year global mortality times compared to those without VTE: 2.21 [95% CI (1.65–2.96), P<.001], 2.04 [95% CI (1.52–2.73), P<.001] respectively.
Discussion
Our nationwide population-based cohort study aimed to investigate the risk of venous thromboembolism (VTE) following esophageal and gastric cancer surgeries, using data from the French National Health Insurance System Database (SNDS), which includes more than 8000 patients. Our findings underscore the significance of preoperative risk assessment in the context of ECS and GCS for cancer, revealing a noteworthy incidence within 90 POD and up to 1-year. Interestingly, this late risk appeared to be more prominent in the ECS group. Furthermore, our study identified 90 POD risk factors for VTE, such as esophageal surgery (compared to gastric surgery), obesity, and a history of VTE. Surprisingly, the presence of preoperative treatment (chemotherapy or radiochemotherapy) was not found, despite existing literature that may suggest otherwise1,14,15.
Many studies have focused on the risks of VTE after complex digestive surgery. Indeed, by applying the Caprini score16, these surgeries are potentially high-risk procedures: patients suffering from cancer, extensive surgery, prolonged surgery duration, and sometimes complicated postoperative recoveries requiring extended bed rest. In this regard, ECS and GCS accumulates these risk factors, to which a significant rate of obese patients (EGJ surgery) and elderly patients can be added17.
Thus, in a national study of American patients operated between 1999 and 2009, Trinh et al. found an inhospital VTE rate of 4.0% after complex oncological surgery (including gynecological and urological surgeries). The authors reported a 10-year increase in inhospital VTE risk, accompanied by a decrease in associated mortality18. In 2011, Merkow et al. found a 30-POD VTE risk of 1.6% after complex oncological surgery, with a higher rate (4.2%) in the subgroup of esophago-gastric surgery (N=2093), in a study derived from an analysis of the NSQIP (National Surgical Quality Improvement Program) database. The authors identified pancreatic surgery as the second highest-risk type of digestive surgery19. These findings are supported by a recent meta-analysis20. In an ancillary analysis of a randomized trial assessing the postoperative addition of low molecular weight heparin (LWMH) after GCS, Jung et al. found 30-POD VTE rates ranging from 3.6 to 0.6%21. However, this was a single-center study involving 682 patients from an Asian population with a lower preoperative BMI (23.7±2.7 kg/m²) compared to our national series (8.4% obese patients) and patients with preoperative treatment were excluded. In a historical cohort (2000–2010) of patients operated on for ECS in Europe, the VTE risk was reported at 2.9%, but this referred to VTE diagnosed during the initial hospital stay; moreover, the study relied on retrospective data22. With 90 POD VTE rates of 6.2 (ECS) and 3.6% (GCS), our data are slightly higher than those reported in the literature. However, our series identified this risk at 90 POD (versus 30-POD), including VTE needing readmissions or treated on an outpatient basis after discharge. In a study using the Nationwide Readmission Database, Mallick et al. 3 found 90-day postoperative readmission rates for VTE of 1.6 (ECS) and 1.1% (GCS). These rates closely resemble those in our French cohort (0.93 and 0.97%, respectively). Few nationwide studies have focused on the late postoperative risk of VTE. Mallick et al. identified a 180-day VTE-related readmission rate of 2.5 (ECS) and 2.3% (GCS). In our study, excluding 90-day VTE cases, we observed readmission rates of 3.0 and 2.98% at 1-year, respectively. More broadly, we identified a risk between 90 days and 1-year postoperation of 7.7 and 5.7%. This significant delayed risk must be acknowledged. However, our results cannot be directly compared to the literature due to our study’s inclusion of both readmitted patients and those treated on an outpatient basis without requiring readmission, a distinction from Mallick et al.’s study3.
Regarding preoperative risk factors for postoperative VTE (excluding cancer location), the literature has identified several factors such as advanced age18–20, obesity, and prolonged operative durations19,20. Concerning the presence of preoperative treatment, the literature is inconclusive, with some studies identifying such treatment as a risk factor20,23 while others do not22. Our study underscores the significance of a history of VTE as a risk factor for VTE and highlights the elevated risk among patients undergoing ECS compared to those undergoing GCS.
The main strength of our study was its being based on the comprehensive French nationwide inpatient database, with more than 8000 patients included followed-up for up to 1-year after surgery. This database has already been used in the field of digestive surgery9–11. Hence, ours is among the major studies in the specific field of VTE risk after ECS and GCS with the added benefit of extending follow-up to 1-year postoperatively. In France, all hospital stays are reimbursed using a Diagnosis-Related Group (DRG) model. Thus, all information (medical information with an ICD-10 classification and French surgical procedure classification) recorded during a hospital stay is available. These data are also available for low-income patients for whom the copayment is waved by French health insurance. Because, we extended our data extraction to outpatient data and with a follow-up of 1-year, we were able to assess not only inpatient but also outpatient VTE, unlike other nationwide studies3,18. Our extensive database avoided any loss to follow-up, which could have led to an underestimation of this risk.
Our study also had several limitations. The first is inherent to studies using medico-administrative databases. Certain important clinical data (for example, weight), laboratory test results, radiological examinations, and anatomopathological findings are not recorded or available in the SNDS database. Hence, we were unable to include the TNM stage according to the TNM Classification of Malignant Tumors Eighth Edition in our multivariable analysis. Furthermore, we did not investigate the impact of adjuvant treatment (chemotherapy/radiochemotherapy) since these treatments often begin in a delayed fashion following such extensive surgeries and typically conclude after the third postoperative month. It is also noteworthy that the very nature of our database does not allow for an exploration of a link between complications of a central venous line (infection/thrombosis) and the risk of VTE. Lastly, we faced challenges in conducting a meaningful analysis of LMWH consumption posthospital discharge. The existing system for drug reimbursement recording in France does not provide insights into daily dosages or the prescribed duration.
Although effective prophylaxis are available, postoperative courses are often conducive to the risk of VTE, especially within 30–90 POD after complex digestive surgery3,9. Prevention is mainly based on the use of LMWH1,7,9. However, because of the increased risk after oncologic digestive surgery, the recommended duration after laparotomy should be extended to 4 weeks1,7. For laparoscopic surgery, international recommendations differ between an extended duration of 4 weeks7 or a shorter duration1. It is worth noting that the use of mechanical methods is no longer recommended as a single treatment7. In all scenarios, the highest prophylactic dose should be used. Our study found a significant incidence of VTE at 1-year, which does not seem to decrease significantly after 90 POD, in patients who in many cases required postoperative systemic anticancer treatment. Considering that 5–10% of patients requiring such treatment will develop VTE24, the question may arise whether to extend the duration of prophylactic treatment in patients benefiting from ECS/GCS and requiring such systemic treatment. Finally, the use of direct oral anticoagulants offers new perspectives in primary prevention, given their greater ease of use compared to LMWH postoperatively, but the literature on this topic remains limited6. Their usefulness could be discussed for patients requiring prophylaxis beyond the usual 30 days6, as demonstrated in the prevention of the risk of recurrence in a randomized trial in nonoperated patients25.
Conclusion
Cancer-related thrombosis, especially following complex surgeries like ECS and GCS, remains a significant concern in oncology care. Our comprehensive study, utilizing the extensive French nationwide inpatient database, has highlighted the increased risk of VTE within 90 POD and extending up to 1-year. Notably, the ECS group exhibited a pronounced late risk. These findings emphasize the crucial importance of VTE prevention and management in this patient cohort. While our study draws attention to the heightened risk of global mortality within 1-year in patients experiencing early or late VTE, it also suggests that future research should delve deeper into the potential advantages of prolonged prophylactic treatments. This is particularly pertinent in the context of postoperative systemic anticancer therapies.
Ethical approval
Analyses of the SNDS database for the present study received the ethical and methodological approval of the French Health Data Hub (Entrepot des données de santé ) and the French Expert Committee for Research, Studies and Evaluations in the Field of Health (Comité d’Expertise pour les Recherches, les Études et les Évaluations dans le domaine de la Santé, N° T32257572020122). Finally, this study was authorized by the French personal data protection authority (Commission Nationale Informatique et Libertés, CNIL, N° 921140).
Consent
No informed patient consent was required because this database is anonymous and all analyses were performed on the devoted secure portal of the French National Health Insurance.
Sources of funding
INNOVEO – The public sponsorship fund of the Brest University Hospital Number of words JT had full access to all study data via Qualees company and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Author contribution
J.T., K.L., F.C., C.L., J.P.M., A.C., B.F., and B.B.: conceptualization; B.F. and J.T.: data curation; J.T. and B.B.: formal analysis; J.T., K.L., and F.C.: funding acquisition; J.T., K.L., B.B., C.L., A.C., and J.P.M.: investigation; J.T., K.L., B.F., and B.B.: methodology; J.T., K.L., and F.C.: project administration; J.T., K.L., B.F., A.C., and B.B.: resources; J.T. and Qualees company: software; J.T., B.B., F.C., and J.P.M.: supervision; J.T., B.B., B.F., A.C., C.L., K.L., J.P.M., and F.C.: validation and writing – original draft.
Conflicts of interest disclosure
All authors have completed the disclosure form.
Research registration unique identifying number (UIN)
Analyses of the SNDS database for the present study received the ethical and methodological approval of the French Health Data Hub (Entrepot des données de santé ) and the French Expert Committee for Research, Studies and Evaluations in the Field of Health (Comité d’Expertise pour les Recherches, les Études et les Évaluations dans le domaine de la Santé, N° T32257572020122). Finally, this study was authorized by the French personal data protection authority (Commission Nationale Informatique et Libertés, CNIL, N° 921140).
Guarantor
Pr Jeremie Thereaux.
Data availability statement
In response to your inquiry regarding data availability, I would like to clarify that access to the French SNDS (Système National des Données de Santé) data is restricted. Obtaining these data requires a specific application to the Health Data Hub (HDH) of France. Due to these constraints, direct public access to the raw dataset used in our study is not feasible. However, interested researchers can apply for access to the SNDS data through the official channels provided by the Health Data Hub.
Provenance and peer review
None.
Previous presentation
All authors gave their final approval of the version to be published.
Supplementary Material
Footnotes
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal's website, www.lww.com/international-journal-of-surgery.
Published online 11 December 2023
Contributor Information
Jérémie Thereaux, Email: jeremie.thereaux@chu-brest.fr.
Bogdan Badic, Email: bogdan.badic@chu-brest.fr.
Basile Fuchs, Email: basile.fuchs@chu-brest.fr.
Cyril Leven, Email: cyril.leven@chu-brest.fr.
Anais Caillard, Email: anais.caillard@chu-brest.fr.
Karin Lacut, Email: jeremie.thereaux@hotmail.fr.
Jean-Philippe Metges, Email: jean-philippe.metges@chu-brest.fr.
Francis Couturaud, Email: francis.couturaud@chu-brest.fr.
References
- 1. Falanga A, Ay C, Di Nisio M, et al. Venous thromboembolism in cancer patients: ESMO clinical practice guideline. Ann Oncol 2023;34:452–467. [DOI] [PubMed] [Google Scholar]
- 2. Fitzmaurice C, Abate D, Abbasi N, et al. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2017: a systematic analysis for the global burden of disease study. JAMA Oncol 2019;5:1749–1768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Mallick S, Aiken T, Varley P, et al. Readmissions from venous thromboembolism after complex cancer surgery. JAMA Surg 2022;157:312–320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Kelly RJ, Ajani JA, Kuzdzal J, et al. Adjuvant nivolumab in resected esophageal or gastroesophageal junction cancer. N Engl J Med 2021;384:1191–1203. [DOI] [PubMed] [Google Scholar]
- 5. Al-Batran SE, Homann N, Pauligk C, et al. Perioperative chemotherapy with fluorouracil plus leucovorin, oxaliplatin, and docetaxel versus fluorouracil or capecitabine plus cisplatin and epirubicin for locally advanced, resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4): a randomised, phase 2/3 trial. Lancet 2019;393:1948–1957. [DOI] [PubMed] [Google Scholar]
- 6. Key NS, Khorana AA, Kuderer NM, et al. Venous thromboembolism prophylaxis and treatment in patients with cancer: ASCO guideline update. J Clin Oncol 2023;41:3063–3071. [DOI] [PubMed] [Google Scholar]
- 7. Farge D, Frere C, Connors JM, et al. 2022 International Clinical Practice Guidelines for the treatment and prophylaxis of venous thromboembolism in patients with cancer, including patients with COVID-19. Lancet Oncol 2022;23:e334–e347. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Hanna NM, Williams E, Kong W, et al. Incidence, timing, and outcomes of venous thromboembolism in patients undergoing surgery for esophagogastric cancer: a population-based cohort study. Ann Surg Oncol 2022;29:4393. [DOI] [PubMed] [Google Scholar]
- 9. Thereaux J, Lesuffleur T, Czernichow S, et al. To what extent does posthospital discharge chemoprophylaxis prevent venous thromboembolism after bariatric surgery?: results from a nationwide cohort of more than 110,000 patients. Ann Surg 2018;267:727–733. [DOI] [PubMed] [Google Scholar]
- 10. Thereaux J, Lesuffleur T, Czernichow S, et al. Association between bariatric surgery and rates of continuation, discontinuation, or initiation of antidiabetes treatment 6 years later. JAMA Surg 2018;153:526–533. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Thereaux J, Lesuffleur T, Czernichow S, et al. Long-term adverse events after sleeve gastrectomy or gastric bypass: a 7-year nationwide, observational, population-based, cohort study. Lancet Diabetes Endocrinol 2019;7:786–795. [DOI] [PubMed] [Google Scholar]
- 12. Mathew G, Agha R, Albrecht J, et al. STROCSS 2021: strengthening the reporting of cohort, cross-sectional and case-control studies in surgery. Int J Surg 2021;96:106165. [DOI] [PubMed] [Google Scholar]
- 13. Bannay A, Chaignot C, Blotière PO, et al. The best use of the Charlson Comorbidity Index with electronic health care database to predict mortality. Med Care 2016;54:188–194. [DOI] [PubMed] [Google Scholar]
- 14. Mulder FI, Horváth-Puhó E, van Es N, et al. Venous thromboembolism in cancer patients: a population-based cohort study. Blood 2021;137:1959–1969. [DOI] [PubMed] [Google Scholar]
- 15. Falanga A, Schieppati F, Russo L. Pathophysiology 1. Mechanisms of Thrombosis in Cancer Patients. Cancer Treat Res 2019;179:11–36. [DOI] [PubMed] [Google Scholar]
- 16. Lobastov K, Urbanek T, Stepanov E, et al. The thresholds of caprini score associated with increased risk of venous thromboembolism across different specialties: a systematic review. Ann Surg 2023;277:929–937. [DOI] [PubMed] [Google Scholar]
- 17. Low DE, Kuppusamy MK, Alderson D, et al. Benchmarking complications associated with esophagectomy. Ann Surg 2019;269:291–298. [DOI] [PubMed] [Google Scholar]
- 18. Trinh VQ, Karakiewicz PI, Sammon J, et al. Venous thromboembolism after major cancer surgery: temporal trends and patterns of care. JAMA Surg 2014;149:43–49. [DOI] [PubMed] [Google Scholar]
- 19. Merkow RP, Bilimoria KY, McCarter MD, et al. Post-discharge venous thromboembolism after cancer surgery: extending the case for extended prophylaxis. Ann Surg 2011;254:131–137. [DOI] [PubMed] [Google Scholar]
- 20. Li M, Guo Q, Hu W. Incidence, risk factors, and outcomes of venous thromboembolism after oncologic surgery: a systematic review and meta-analysis. Thromb Res 2019;173:48–56. [DOI] [PubMed] [Google Scholar]
- 21. Jung YJ, Seo HS, Park CH, et al. Venous thromboembolism incidence and prophylaxis use after gastrectomy among korean patients with gastric adenocarcinoma: the PROTECTOR randomized clinical trial. JAMA Surg 2018;153:939–946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Mantziari S, Gronnier C, Pasquer A, et al. Incidence and risk factors related to symptomatic venous thromboembolic events after esophagectomy for cancer. Ann Thorac Surg 2016;102:979–984. [DOI] [PubMed] [Google Scholar]
- 23. Fehr M, Hawle H, Hayoz S, et al. High thromboembolic event rate in patients with locally advanced oesophageal cancer during neoadjuvant therapy. An exploratory analysis of the prospective, randomised intergroup phase III trial SAKK 75/08. BMC Cancer 2020;20:166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Di Nisio M, Porreca E, Candeloro M, et al. Primary prophylaxis for venous thromboembolism in ambulatory cancer patients receiving chemotherapy. Cochrane Database Syst Rev 2016;12:Cd008500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Schrag D, Uno H, Rosovsky R, et al. Direct oral anticoagulants vs low-molecular-weight heparin and recurrent VTE in patients with cancer: a randomized clinical trial. Jama 2023;329:1924–1933. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
In response to your inquiry regarding data availability, I would like to clarify that access to the French SNDS (Système National des Données de Santé) data is restricted. Obtaining these data requires a specific application to the Health Data Hub (HDH) of France. Due to these constraints, direct public access to the raw dataset used in our study is not feasible. However, interested researchers can apply for access to the SNDS data through the official channels provided by the Health Data Hub.