Clinical outcomes of venous tumor thrombectomy in renal cell carcinoma

Abstract

Objective

Renal cell carcinoma (RCC) has a unique feature in which tumor thrombus extends through the vein into the inferior vena cava. RCC with venous tumor thrombus is classified based on the level of thrombus. Standard treatment involves a combination of radical nephrectomy and tumor thrombectomy, despite the technical challenges associated with this approach. We aimed to evaluate the clinical outcomes based on the tumor thrombus level.

Methods

We conducted a retrospective analysis of patients who underwent tumor resection and thrombectomy for RCC with venous tumor thrombus at a single center between January 2001 and December 2022. Tumor thrombus levels were classified into four levels (I-IV) according to the Mayo Clinic grading system, and patients were grouped based on the hepatic vein level: levels 0, I, and II were categorized as group 1, while levels III and IV were categorized as group 2. Survival outcomes were compared between the two groups using Kaplan-Meier survival analysis.

Results

A total of 287 patients with RCC with venous tumor thrombus who underwent tumor resection and thrombectomy were identified. The distribution of patients according to the tumor thrombus level was as follows: 84 patients had level 0 thrombus, 71 had level I, 85 had level II, 19 had level III, and 28 had level IV. This results in 240 patients in group 1 (levels 0, I, and II) and 47 patients in group 2 (levels III and IV). The 5-year survival rates were 46.7% for group 1 and 53.2% for group 2. Kaplan-Meier survival analysis showed no significant difference in 5-year overall survival between the two groups (P = .75). However, patients in group 2 had a significantly longer postoperative hospital stay (21.43 ± 23.09 days) compared with those in group 1 (10.63 ± 6.31 days; P = .004).

Conclusions

Tumor thrombus level is not a significant prognostic factor in RCC. Aggressive surgical removal should be considered, even for thrombus extending beyond the hepatic vein, given the acceptable prognosis.

Article Highlights.

• •Type of Research: Single-center retrospective cohort study
• •Key Findings: Among 287 renal cell carcinoma patients with venous tumor thrombus who underwent surgery, 240 had thrombus level 0, I, or II (group 1) and 47 had thrombus level III or IV (group 2). The 5-year survival rates did not differ significantly between the groups (46.7% vs. 53.2%).
• •Take Home Message: Despite the technical challenges of surgery for higher tumor thrombus levels, surgical resection can be considered, because the survival outcomes are not affected by the tumor thrombus level significantly.

Renal cell carcinoma (RCC) has a unique biological feature of tumor thrombus extending through the vein into the inferior vena cava (IVC), making treatment challenging. Tumor extension into the renal vein and IVC occurs in up to 4% to 23% of RCC cases.¹ RCC with venous tumor thrombus can be classified into four levels (levels I-IV) based on the extent of the thrombus according to the Mayo Clinic grading system.²

The standard treatment for RCC with venous tumor thrombus involves radical nephrectomy combined with tumor thrombectomy, despite the technical challenges associated with this approach.^3,4 Before 2000, surgical outcomes for patients with RCC and IVC tumor thrombus were generally poor.^5,6 However, advances in surgical techniques, such as improved preoperative imaging, intraoperative transesophageal echocardiography, development of refined vascular surgical instruments, accumulation of surgical experience, and multidisciplinary collaboration, have markedly improved the outcomes for these patients undergoing radical nephrectomy and tumor thrombectomy.⁷ Therefore, curative-intent surgical resection should be considered even in patients with high-level thrombus.

Given that the technical difficulties and potential postoperative morbidity increase with the level of tumor thrombus, higher levels may result in greater postoperative mortality and longer recovery periods. Additionally, whereas some studies have identified various prognostic factors for RCC with tumor thrombus, conflicting reports persist on whether the level of tumor thrombus affects long-term survival.^6,8

In this study, we aimed to evaluate the clinical outcomes based on the tumor thrombus level to determine whether higher levels affect the short-term and long-term outcomes in patients with RCC who underwent radical nephrectomy with tumor thrombectomy.

Methods

Patients

This single-center, retrospective observational study analyzed data from the medical records of patients who underwent radical nephrectomy combined with tumor thrombectomy to treat RCC with venous tumor thrombus at Asan Medical Center from January 2001 to December 2022. Patients were included regardless of prior treatments or the presence of metastasis. Additionally, only those with RCC confirmed by histology after surgery were included. Because the presence of venous tumor thrombus classifies RCC as stage T3a or higher under the Tumor Node Metastasis (TNM) staging system, patients with stage T1 or T2 tumors were inherently excluded. The study protocol was approved by the Institutional Review Board of Asan Medical Center, Republic of Korea, which waived the requirement for informed patient consent owing to the study's retrospective design. All medical records were reviewed by one independent reviewer.

Treatment

This study included patients who underwent radical nephrectomy with tumor thrombectomy. Surgery was performed either immediately after diagnosis or after targeted therapy. Nephrectomy in the presence of metastatic disease was considered when it was feasible for cytoreduction and/or alleviation of symptoms. Decisions regarding surgical treatment were made through a multidisciplinary team approach involving urologists, vascular surgeons, oncologists, and anesthesiologists, based on each patient's clinical status and expected surgical benefit. Radical nephrectomy was performed using either open or laparoscopic techniques, with all open surgeries performed via a transperitoneal approach. All cases requiring IVC venotomy were performed using the open method. Routine anticoagulation prior to surgery was not administered, nor was renal artery embolization conducted before surgery. Intraoperative transesophageal echocardiography was used in some cases involving IVC tumor thrombus to detect the occurrence of embolism and monitor cardiac function after IVC clamping. When clamping above the hepatic vein level was necessary, the Pringle maneuver was performed to reduce blood flow from the hepatic veins to the IVC, along with sequential clamping to minimize supra hepatic clamping time. For thrombus extending above the diaphragm, sternotomy and cardiopulmonary bypass (CPB) were used in some cases.

Postoperative follow-up was conducted in accordance with the National Comprehensive Cancer Network guidelines for RCC, with imaging surveillance performed at intervals recommended by these guidelines.

Definitions

The Mayo classification system was used to categorize the level of tumor thrombus as follows: level 0 (thrombus limited to the renal vein), level I (thrombus extending into the IVC within 2 cm above the renal vein level), level II (thrombus extending into the IVC >2 cm above the renal vein level), level III (thrombus at the level of or above the hepatic veins but below the diaphragm), and level IV (thrombus extending above the diaphragm).² The level of tumor thrombus was determined by integrating operative findings, preoperative computed tomography findings, and pathological reports.

Based on the anatomical location of the tumor thrombus in relation to the hepatic veins, patients were further stratified into two groups for analysis: group 1 included patients with thrombus below the hepatic veins (Mayo levels 0, I, and II), and group 2 included patients with thrombus above the hepatic veins (Mayo levels III and IV). This classification reflects differences in surgical complexity. Tumor thrombi above the hepatic veins often require more extensive surgical maneuvers, including high-level IVC clamping, liver mobilization, and, in some cases, application of the Pringle maneuver to reduce hepatic inflow during suprahepatic IVC clamping.

Study end points

The primary objective of this study was to evaluate the impact of tumor thrombus level on long-term survival outcomes after tumor thrombectomy combined with radical nephrectomy. The secondary objective was to assess short-term survival outcomes and hospital stay durations based on tumor thrombus level.

Statistical analyses

Categorical variables were presented as frequencies or percentages and compared using the χ² or Fisher's exact test, as appropriate. Normally distributed continuous variables were presented as means with standard deviations and compared using the Student t tests, whereas non-normally distributed continuous variables were presented as medians with ranges and compared using Mann-Whitney rank test variables. Variables significantly associated with study outcomes were identified using Cox proportional hazards regression models, with the results reported as hazard ratios (HRs) with 95% confidence intervals (CIs). Variables with P values of less than .1 in univariable analyses were included in multivariable analysis. The 5-year survival rates of the patients, grouped according to tumor thrombus level, were calculated using the Kaplan-Meier method and compared using log-rank tests. All statistical analyses were performed using SPSS Statistics for Windows, version 21.0 (IBM Corp.), with P values of less than .05 defined as statistically significant.

Results

Between January 2001 and December 2022, 287 patients with RCC with venous tumor thrombus underwent radical nephrectomy and tumor thrombectomy at Asan Medical Center, Seoul, Korea. The distribution of patients according to tumor thrombus level included 84 patients with level 0 thrombus, 71 with level I, 85 with level II, 19 with level III, and 28 with level IV. Patients were categorized based on the tumor thrombus level: thrombus levels below the hepatic veins (levels 0, I, and II) were defined as group 1 (n = 240), and thrombus levels above the hepatic veins (levels III and IV) were defined as group 2 (n = 47) (Fig 1).

Fig 1.

Flow chart of this study. RCC, renal cell carcinoma.

Upon comparison of the baseline clinicopathological characteristics between the two groups, no significant differences were found in age, sex, comorbidities, or N stage. However, significant differences were noted in T stage, M stage, and histologic type. Group 2 had a higher prevalence of T3bc stages (44.6% vs 70.2%; P < .001) and non-clear cell type cancers (15.8% vs 29.8%; P = .039), and group 1 had a greater proportion of M1 stages (32.1% vs 10.6%; P = .002) (Table I).

Table I.

Baseline clinicopathological characteristics of the patients

	Group 1 (n = 240)	Group 2 (n = 47)	Total (n = 287)	P value
Sex				.48
Male	188 (78.3)	34 (72.3)	222 (77.4)
Female	52 (21.7)	13 (27.7)	65 (22.6)
Age at operation, years	58.28 $\pm$ 11.18	59.15 ± 10.77	58.42 ± 11.1	.62
Comorbidities
HTN	116 (48.3)	26 (55.3)	142 (49.5)	.38
DM	58 (24.2)	7 (14.9)	65 (22.6)	.17
T stage				<.001
T3a	107 (44.6)	5 (10.6)	112 (39.0)
T3bc	107 (44.6)	33 (70.2)	140 (48.8)
T4	26 (10.8)	9 (19.1)	35 (12.2)
N stage				.6
N0	139 (57.9)	30 (63.8)	169 (58.9)
N1	50 (20.8)	10 (21.3)	60 (20.9)
Nx	51 (21.2)	7 (14.9)	58 (20.2)
M stage				.002
M0	150 (62.5)	35 (74.5)	185 (71.8)
M1	77 (32.1)	5 (10.6)	82 (28.2)
Mx	13 (5.4)	7 (14.9)	20 (7.0)
Histological type				.039
Clear cell	202 (84.2)	33 (70.2)	235 (81.9)
Non-clear cell	38 (15.8)	14 (29.8)	52 (18.1)

DM, Diabetes mellitus; HTB, hypertension.

Values are number (%) or mean ± standard deviation.

Among the 287 patients included in the study, 10 patients (3.5%) underwent radical nephrectomy using a laparoscopic approach, all of whom had level 0 tumor thrombus. The remaining 277 patients (96.5%) underwent open surgery. No cases were performed using robotic-assisted surgery. The median follow-up duration was 34 months (range, 0-257 months; interquartile range, 14.5-75.0 months).

Among the 84 patients with level 0 tumor thrombus, 78 patients underwent renal vein dissection and ligation without requiring IVC clamping or thrombectomy. For the remaining 6 patients, the thrombus was located close to the IVC, necessitating partial IVC clamping and IVC repair to remove the thrombus safely.

Among the patients with level III or IV thrombus, CPB was used in 18 patients (38.3%) and sternotomy was performed in 23 patients (48.9%). A subgroup analysis of patients with level III or IV thrombus (group 2) according to CPB use was performed. Although not statistically significant, patients who underwent surgery with CPB showed a trend toward longer intensive care unit stays (8.55 vs 6.04 days; P = .58), longer hospital stays (26.86 vs 16.2 days; P = .12), and a higher rate of perioperative complications (31.8% vs 12.0%; P = .15), compared with those treated without CPB.

The 5-year all-cause mortality rates were not significantly different between the two groups (53.3% vs 46.8%; P = .41) (Table II). Similarly, the Kaplan-Meier analysis of 5-year overall survival showed no significant differences between the two groups (Fig 2).

Table II.

Clinical outcomes according to the tumor thrombus level

	Group 1 (n = 240)	Group 2 (n = 47)	Total (n = 287)	P value
Hospital stay, days	10.63 ± 6.31	21.43 ± 23.09	12.28 ± 11.36	.004
ICU stay, days	1.15 ± 2.93	7.21 ± 15.24	2.14 ± 7.04	.009
Perioperative complications (Clavien-Dindo grade ≥3)	5.8	21.3	8.4	.002
In hospital mortality	0.4	2.1	0.7	.28
3-Months mortality	8.2	6.8	8.0	>.99
5-Year mortality	53.3	46.8	52.3	.41

Values are mean ± standard deviation or percent.

Fig 2.

Kaplan-Meier curve of 5-year survival in renal cell carcinoma (RCC) patients stratified by venous thrombus level. The P value shown represents the result of the log-rank test comparing the entire survival curves.

To account for the higher prevalence of M1 disease in group 1, a Kaplan-Meier analysis was conducted on patients without metastatic disease. After excluding these patients, the analysis still showed no significant difference in 5-year overall survival between the two groups (Fig 3).

Fig 3.

Kaplan-Meier curve of 5-year survival in renal cell carcinoma (RCC) patients without metastatic disease, stratified by venous thrombus level. The P value shown represents the result of the log-rank test comparing the entire survival curves.

In the Cox regression analysis to determine whether the level of tumor thrombus independently affects overall survival, it was found that the group stratified by tumor thrombus level did not significantly impact overall survival (HR, 0.95; 95% CI, 0.6-1.45; P = .81). However, in the multivariate analysis, T3bc stage (reference: T3a stage, HR, 2.63; 95% CI, 1.73-3.99; P < .001), N1 stage (reference: N0 stage, HR, 2.33; 95% CI, 1.62-3.35; P < .001), and M1 stage (reference: M0 and Mx stage, HR, 1.98; 95% CI, 1.41-2.79; P < .001) were found to decrease overall survival significantly (Table III).

Table III.

Factors associated with 5-year overall survival during follow-up

	Univariable analysis		Multivariable analysis
	HR (95% CI)	P value	HR (95% CI)	P value
Age at operation, years	0.99 (0.97-1.0)	.15
Sex (Ref: female)	0.92 (0.63-1.35)	.66
TNM staging
T stage (Ref: T3)	3.28 (2.18-4.93)	<.001	2.63 (1.73-3.99)	<.001
N stage (Ref: N0)	2.87 (2.02-4.07)	<.001	2.33 (1.62-3.35)	<.001
M stage (Ref: M0)	2.39 (1.72-3.34)	<.001	1.98 (1.41-2.79)	<.001
Histological type (Ref: clear cell RCC)	1.82 (1.24-2.67)	.002	1.44 (0.97-2.12)	.07
Thrombus level (Ref: group 1)	0.95 (0.6-1.49)	.81

CI, Confidence interval; HR, hazard ratio.

To assess the short-term clinical outcomes of the operation, unadjusted comparisons between the two groups were performed for 3-month mortality, in-hospital mortality, and hospital stay. There were no significant differences between the two groups in terms of 3-month mortality (8.2% vs 6.8%; P > .99) or in-hospital mortality (0.4% vs 2.1%; P = .28). However, the hospital stay was significantly longer in group 2 (10.63 ± 6.31 days vs 21.43 ± 23.09 days; P = .004). The intensive care unit stay was also longer in group 2 (1.15 ± 2.93 days vs 7.21 ± 15.24 days; P = .009). Additionally, perioperative complications of Clavien-Dindo grade 3 or higher were significantly more frequent in group 2 (5.8% vs 21.3%; P = .002) (Table II).

Discussion

In this study, we found that the 5-year overall survival rate of patients with RCC with tumor thrombus who underwent radical nephrectomy with tumor thrombectomy did not differ significantly according to the tumor thrombus level. Cox regression analysis also demonstrated that tumor thrombus level did not affect the overall survival rate significantly. Similarly, short-term clinical outcomes, including in-hospital mortality and 3-month mortality, were not significantly different between the groups. However, hospital stay was significantly longer in group 2 patients.

The tumor thrombus extending through the vein into the IVC can occur in a few types of cancer; however, in adults, it is most common in RCC.⁹ Other malignancies that may exhibit tumor thrombus extension into the IVC include hepatocellular carcinoma, Wilms tumor, and adrenocortical carcinoma.10, 11, 12 When tumor thrombus is present, tumor thrombectomy should be performed in combination with radical nephrectomy, unless it is confined to a small portion of the renal vein. This tumor thrombectomy procedure is complex and challenging, and the tumor thrombus level is important in determining the surgical approach.¹³ Owing to the complexity of operating on high-level tumor thrombus, complication rates tend to increase with the level of thrombus.^14,15 Surgery for level III or IV thrombus, which extends beyond the hepatic vein level, requires high-level vena cava clamping and extensive tissue dissection.¹⁶

The impact of tumor thrombus level on the clinical outcomes of patients who undergo surgery is controversial.⁸ Several retrospective studies, including those by Blute et al,⁶ Shiff et al,¹⁷ and a Mayo Clinic study by Lambert et al,¹⁸ have shown that cancer-specific survival is not significantly affected by the level of IVC tumor thrombus.

However, several studies have suggested that the tumor thrombus level is associated with clinical outcomes, including patient survival. For example, a retrospective cohort study by Haferkamp et al,⁸ which analyzed 134 patients, reported that patients with level III or IV thrombus had significantly worse cancer-specific survival compared with those with level I or II thrombus. A systematic review and meta-analysis by Gu et al,¹⁹ which included 35 retrospective studies with a total of 11,929 patients, also demonstrated that IVC tumor thrombus was associated with poorer cancer-specific survival compared with renal vein-limited thrombus.

Also, a retrospective study by Zapała et al²⁰ from Poland, which included 142 patients, indicated that, although the level of tumor thrombus does not affect long-term survival, it does impact 1-year survival rates. This result may be attributed to the greater likelihood of perioperative complications in patients with high-level tumor thrombus. The inconsistency between studies may also stem from differences in surgical techniques, surgical outcomes, patient populations included in the studies, follow-up duration, and the clinicopathological factors assessed in combination with the level of tumor thrombus. Before 2000, surgical outcomes for patients with RCC with IVC tumor thrombus were generally poor, with reported 5-year survival rate ranging from approximately 26% to 34%.^5,6 However, with advances in surgical techniques, outcomes have improved. Relatively recent studies have reported a 5-year overall survival rate of 51.0%, which is similar to our result of 47.7%.⁷

In this study, both short-term and long-term outcomes did not differ between the groups stratified by the level of tumor thrombus. The results for long-term outcomes were consistent with previous studies; however, the short-term outcomes did not align with our initial assumption that patients with higher-level tumor thrombus would have worse early postoperative outcomes owing to more complex surgeries and higher perioperative risks. Considering the result that the hospital stay for group 2 was significantly longer than for group 1, it can be inferred that patients with high-level tumor thrombus experienced more complex surgeries and longer recovery periods. A retrospective study by Dell’Oglio et al²¹ reported that patients with level III or IV tumor thrombus who underwent surgery for RCC required more intraoperative blood transfusions and had a higher incidence of postoperative complications classified as Clavien-Dindo grade 3 or higher, compared with those with level I or II thrombus. However, these factors did not impact in-hospital mortality or 3-month mortality, likely owing to advances in surgical techniques and optimal postoperative management. A previous study by Blute et al,⁶ which included 540 patients, demonstrated a temporal trend toward reduced early surgical complications and operative mortality over time. In their analysis, the incidence of early complications decreased from 13.4% (1970-1989) to 8.1% (1990-2000), and operative mortality decreased from 3.8% to 2.0% overall. Although these differences did not attain statistical significance, they may reflect improvements in surgical technique and perioperative management over the decades.⁶ Additionally, at our center, CPB and sternotomy were not universally required for level III or IV patients, because thrombectomy could be accomplished without them in selected cases, depending on the extent of thrombus, even when the thrombus extended above the diaphragm.^16,22,23 In this study, among the patients with level III or IV thrombus, CPB was used in 18 patients (38.3%) and sternotomy was performed in 23 patients (48.9%). For thrombectomy of level III or IV thrombus without using CPB and sternotomy, it is crucial to isolate the IVC completely from the central tendon of the diaphragm. Once the supradiaphragmatic IVC is fully exposed, the tumor thrombus must be pulled downward carefully to a position where it can be clamped, with great care taken to avoid embolization.²⁴ Similar to bland thrombus of the IVC, embolization of a floating tumor thrombus can produce fatal outcomes.⁹ Therefore, intraoperative monitoring with transesophageal echocardiography can be helpful during this process. In our cohort, intraoperative embolization occurred in eighty patients (2.8%). Of these, seven patients required sternotomy and CPB to retrieve the embolized thrombus from the heart, and one patient was managed using peripheral CPB via a minithoracotomy approach. All cases were managed successfully without intraoperative mortality. Owing to these techniques, the short-term outcomes, as well as long-term outcomes, were not significantly different between the groups, even though the duration of hospital stay was longer in group 2. It is suggested that reducing the risk of the operation by avoiding sternotomy and tailoring the surgical approach to the individual patient's thrombus location can improve both short-term and long-term outcomes.

This study has several limitations. First, being a single-center, retrospective study, it is subject to biases, and caution is needed when generalizing the results. Second, there is a considerable difference in the number of patients between group 1 and group 2, as well as differences in the proportions of T stage and M stage between the two groups. To address these limitations, we conducted a subgroup analysis excluding M1 stage and used a Cox regression test to analyze the impact of T stage and M stage variables. Thirdly, the long study period (2000-2022) may have introduced heterogeneity in treatment approaches, because advancements in surgical techniques, perioperative care, and systemic therapies over time could have influenced the outcomes.

Conclusions

Our findings suggest that, although the level of tumor thrombus does not affect long-term or short-term survival outcomes, it is associated with a prolonged hospital stay, likely reflecting the increased surgical complexity of a higher-level thrombus. Given the encouraging surgical outcomes even in patients with high-level tumor thrombus, our results support surgical resection in such cases and emphasize the importance of tailoring the surgical approach according to the extent and location of the thrombus.

Author contributions

Conception and design: SL, HJ, YH, YC, JG

Analysis and interpretation: SL, HJ, JG

Data collection: SL, HJ, JG

Writing the article: SL, HJ, JG

Critical revision of the article: SL, HJ, YH, YC, JG

Final approval of the article: SL, HJ, YH, YC, JG

Statistical analysis: SL, HJ

Obtained funding: Not applicable

Overall responsibility: JG

SL and HJ contributed equally to this article and share co-first authorship.

Funding

None.

Disclosures

None.