Venous Thromboembolism (VTE) in Post-prostatectomy Patients- Systematic Review and Meta-Analysis

1. Introduction:

Urological cancer surgical procedures have evolved remarkably over the years, the open surgical methods have now largely been replaced by endoscopic or minimally invasive procedures including laparoscopy and robot-assisted methods. Perioperative venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), represents a serious and potentially fatal complication after urological cancer surgeries [1,2]. DVT incidence, without prophylaxis, has been estimated to be 10-40% among medical and general surgical patients [3]. PE results in approximately 10% of hospital deaths- the most common cause of inpatient deaths [4]. Furthermore, VTE accounts for the most common preventable cause of hospital-related deaths [5,6,7]. National Surgical Quality Improvement Program (NSQIP) data revealed a 2.5-fold increased risk of VTE within 30 days in abdominal procedures compared to procedures such as breast surgery [8]. One of the common abdominal urological cancer surgeries is radical prostatectomy (RP), performed for prostate cancer (PCa), which also evolved from open to robotic-assisted surgery. Ultrasound evaluation of VTE occurrence in post-RP patients is estimated at 20% [9].

Thromboprophylaxis helps in reducing morbidity and mortality in surgical patients, using either mechanical methods to promote venous outflow from the legs or pharmacological prophylaxis (PP) in the form of antithrombotic drugs [10]. The efficacy of thromboprophylaxis in decreasing the incidence of venous thromboembolic events has been demonstrated in randomized controlled clinical trials [11]. Similarly, meta-analyses by various specialties looking into VTE prophylaxis in abdomen/pelvic surgeries concluded that anticoagulants such as low molecular weight heparins (LMWH) decrease the relative risk of VTE by approximately 50%. However, LMWH administration also increases the relative risk of major bleeding by approximately 50% [12,13]. Regarding prophylaxis after RP, there is a lack of high-quality evidence or standard practice across the globe. In the UK, 98% of patients receive PP after RP, while only 17.8% receive PP after RP in the USA [11,14]. One US study found that 30% of patients post-RP received no VTE prophylaxis [14].

The aim of this systematic review was to investigate different aspects of VTE in post-RP patients. The primary aim was to perform a systematic review and meta-analysis (wherever possible) of overall VTE occurrence in post-RP patients as well as VTE occurrence in relation to surgical approach, pelvic lymph node dissection, and type of prophylaxis (mechanical or combined prophylaxis). The secondary aim was to investigate the impact of other risk factors of VTE in post-RP patients.

2. Material and Methods

2.1. Protocol and registration:

This systematic review is registered with the PROSPERO International Registry (CRD42022364222).

2.2. Evidence Acquisition:

The inclusion criteria for studies included the following-

• Population: Male patients, prostatectomy for prostate cancer,
• Intervention: Pharmacological/ Combined Prophylaxis (PP) for VTE
• Comparator/Control: No Prophylaxis or Mechanical Prophylaxis for VTE

The exclusion criteria for studies included the following-

• Population: Prostatectomy for non-prostate cancer or part of other surgery like cystoprostatectomy.
• Intervention: If the interventions are ill-defined or structural methods are inadequate.
• Comparator/ Control: Studies that lacked proper grouping into control, and intervention.
• Study Design: Studies that did not fulfill the above criterion and lacked any defined outcomes.

2.3. Outcome measures:

• Primary Outcomes- VTE occurrence with
  • Overall Incidence of VTE in post-RP patients
  • Surgical approach. (Open, Minimally Invasive)
  • Pelvic Lymph node dissection (PLND)
  • Prophylaxis (No prophylaxis, Mechanical only, Pharmacological only, Combined).
• Secondary Outcomes-

Risk factors for VTE in Post-RP patients including age, body mass index (BMI), smoking, personal or family history of VTE

2.4. Search Methods:

We followed the “Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA)-2020 protocol, for this systematic review [15]. The PRISMA Checklist for this review is shown in Supplementary Table 1. An electronic search was carried out into MEDLINE, EMBASE, and Cochrane Library with results from Cochrane CENTRAL, ClinicalTrials.gov, and the International Clinical Trials Registry Platform (ICTRP). There was no date or language limit. Search terms included ‘Prostatectomy’, ‘Radical’, ‘Cancer’, ‘Venous thromboembolism’, ‘Deep Vein Thrombosis’, ‘Pulmonary Embolism’, ‘Prophylaxis’, ‘Pharmacological’, ‘Mechanical’, and ‘surgical procedures.’ Boolean operators (‘And’/ ‘Or’) were used. The search strategy is attached as Supplementary Table 2.

2.5. Study Selection:

Rayyan software (a free web tool designed to help researchers speed up the process of screening and selecting studies) was used to aid in the process of duplicate removal, and initial screening, and facilitate author collaboration. After deleting duplicates, following the inclusion and exclusion criterion, studies were shortlisted. Full-text papers of the shortlisted studies were reviewed by the SQ3R (Survey, Question, read, recite, review) technique. The process of reviewing studies was completed independently by three authors (M.W, A.M, S.Mu, J.A). In case of any disagreements, the discussion was carried out with the rest of the authors (S.Ma, G.N, B.S) to reach a consensus.

2.6. Data Extraction:

The data was initially divided into two parts. The first part included a detailed description of all the studies included. The second part included grouping data into columns- the number of patients, type of surgical approach procedure, pelvic lymph node dissection, post-op VTE Prophylaxis (Duration/ Type of medication), and additional VTE (Mechanical). Data extraction was performed by two authors independently (M.W, A.M).

2.7. Quality assessment:

For quality assessment of the studies included, the Newcastle-Ottawa Scale (NOS) was used [16].

2.8. Statistical analysis:

It was carried out in three parts-

a.

Statistical evaluation of overall VTE occurrence

The meta-analysis pooled together the results from the different studies, calculating the overall percentage of patients in which the outcomes occurred. Due to the rare occurrence of the outcomes, the Freeman-Tukey double arcsine transformation was performed before analysis. This was used to stabilize the variances when the proportions were close to zero, and a normal approximation to the binomial distribution did not hold.

The DerSimonian-Laird random-effects method was used for the analysis, regardless of the degree of heterogeneity between the study results. Heterogeneity between studies was assessed based on the significance of the between-study heterogeneity and on the size of the I² value. Substantial heterogeneity was assumed if the I² value was above 50%.

b.

Statistical evaluation of VTE occurrence depending on the type of surgical procedure (Open/ Minimally Invasive Surgery (MIS)) and whether PLND was performed or not

Meta-analysis methods were used to pool together the within-study differences in the outcome of different procedures. The outcome of interest was the occurrence of VTE, which was binary in nature. Differences between procedures were expressed as a relative risk.

The DerSimonian-Laird random-effects method was used for the analysis, regardless of the degree of heterogeneity between the study results. Heterogeneity between studies was assessed based on the significance of the between-study heterogeneity and on the size of the I² value. Substantial heterogeneity was assumed if the I² value was above 50%.

Data was collected from a series of studies. However, not all studies collected data from more than one type of procedure, and thus a comparison between procedures was based on data from a more limited number of studies. The data collected enabled two different comparisons to be made:

• MIS procedures vs. Open procedures
• Procedures using PLND vs. procedures without PLND

c.

Statistical Evaluation of VTE occurrence depending on the method of prophylaxis used (mechanical or combined).

The systematic review found only two studies that collected data on both methods, and that could be used in a within-study comparison. However, additional studies collected data on either one method or the other. To be able to include all studies in the analysis, a within-study comparison was not made. Instead, the occurrence of VTE was pooled separately for the mechanical and combined methods. The difference in VTE occurrence between methods was assessed based on the significance of the between-group heterogeneity. Similar statistical methods were used as described already for overall VTE occurrence.

3. Results

3.1. Study selection results:

In the initial search, 394 studies were identified. A total of 18 were removed (Including 15 duplicates). 376 studies were screened. After the title and abstract review, 302 studies were excluded. Full-text retrieval was sought for 74 studies, however, we could retrieve only 72. Out of 72 studies, 41 studies were excluded after a full-text review. 31 remaining studies were scrutinized for eligibility criteria, and in the end, 16 studies were included in the systematic review. The PRISMA-2020, flow diagram is shown in Figure 1.

3.2. Quality assessment results:

The quality assessment results of the included studies are shown in Table 1

3.3. Study characteristics:

All sixteen studies included are summarized in Table 2, highlighting the study type, period of study, main characteristics/ methodology, and conclusions.

3.4. Clinic-pathological results:

The results are summarized in Supplementary Table 3. The main highlight is that surgery was usually performed for T2-stage of prostate cancer (54%).

3.5. Demographics and VTE risk factors

The VTE risk factors including age, BMI, personal history of VTE as well as family VTE are summarized with remarks in Table 3. The age was described in 13 studies, with the mean age of patients undergoing surgery being 64 years (60-72 years). Six studies provided details about BMI. The average BMI was 27.6%. Personal history of VTE had been investigated by 3 studies, with the average being 2.3% and family history of VTE in one study, 4.0%. Three studies investigated smoking and were found in 30% of patients. Lastly, only one study investigated Caprini’s score in their cohort of patients.

3.6. Surgical Procedure results:

Out of 16 studies, 15 have mentioned detailed surgical approaches offered for radical prostatectomy (Open/Laparoscopic/ Robotic). 11 studies have described the results of pelvic lymph node dissection in their cohort of patients. Results are summarized in Table 4.

3.7. Thromboprophylaxis and VTE episodes

The results of different types of thromboprophylaxis as well as the number of VTE episodes are provided in Table 5. Fourteen studies have provided details about VTE prophylaxis (95,052 patients). Of these patients, 22.4% did not receive any form of VTE prophylaxis. 41.8% of patients received mechanical prophylaxis followed by 30% of patients receiving combined (mechanical as well as pharmacological) prophylaxis and lastly, 5.7% of patients received only pharmacological prophylaxis Figure 2.

3.8. Surgical approaches and VTE episodes

Total VTE episodes in open RP were discussed in 11 studies with a mean VTE incidence of 1.5%. 8 studies discussed VTE episodes in MIS having a mean incidence of 0.9%. 8 studies investigated DVT incidence in post-RP patients and this was 0.3%. 13 studies described PE incidence in post-RP patients and had a mean value of 1.5%. These results are tabulated in Table 6.

3.9. Duration and timing of Thromboprophylaxis

In most studies, mechanical prophylaxis was initiated in patients prior to surgery and continued till patient discharge from the hospital. In some studies, patients were also encouraged to ambulate on discharge [20].

However, there is a lack of uniformity in the case of PP in terms of both starting as well as duration. Some studies started PP before surgery and continued 8 hourly till discharge from the hospital [17]. Others emphasized starting PP after surgery and continuing for 4 weeks after [18].

3.10. Statistical Results

a.

Statistical outcome of overall VTE occurrence

A summary of the meta-analysis results is summarized in Table 7. The first figures are the number of studies with valid data included in each analysis. Subsequently, details of the heterogeneity are reported, both in terms of the significance and the I² values. The next figures are the pooled percentage of patients in which the outcome occurred for each method, presented with a corresponding confidence interval. The pooled results suggested that, for all patients combined, the occurrence of VTE was 1.0% (95% CI: 0.5% to 1.5%). The graphical illustration of the meta-analysis results is shown in Figure 3.

b.

Statistical outcome of VTE occurrence depending on the type of surgical procedure (Open/MIS) and whether PLND was performed or not.

A summary of the meta-analysis results is summarized in Table 8. The first figures are the number of studies with valid data included in each analysis. Subsequently, details of the heterogeneity are reported in terms of the significance and the I² values. The final columns give the differences in outcome between the two risk groups. The pooled difference in outcome between groups is expressed as a relative risk. The pooled figure is presented along with a corresponding confidence interval. The final column shows the p-values indicating the significance of the differences between risk groups.

The analysis suggested a statistically significant difference in outcome between MIS and Open procedures, based on data from 5 studies. The pooled results suggested a lesser occurrence of VTEs in MIS procedures compared to Open procedures. The number of VTEs was just over a third lower (0.63 times as large) in the MIS group when compared to the Open Group. There was little heterogeneity between the different studies. A graphical illustration of the meta-analysis results is shown in Figure 4.

There were two studies providing sufficient data to compare patients where PLND was used and not used. The data suggested some evidence that VTE was more common with the use of PLND. However, the result did not quite reach statistical significance. The pooled results suggested that a VTE was 2.8 times more likely when PLND was performed compared to when it wasn’t. There was little heterogeneity between the two studies. A graphical illustration of the results is shown in Figure 5.

c.

Statistical outcome of VTE occurrence depending on the method of prophylaxis used (mechanical or combined).

The meta-analysis pooled together the results from studies with patients undergoing mechanical and combined prophylaxis methods. A summary of the meta-analysis results is summarised in Table 9. The first figures are the number of studies with valid data included in each analysis. Subsequently, details of the heterogeneity are reported in terms of the significance and the I² values. The next figures are the pooled percentage of patients with VTE for each method, presented with a corresponding confidence interval. The final column shows the p-values indicating the significance of the differences in VTE between groups.

The pooled results suggested that 0.7% of patients underwent a VTE for the mechanical method, whilst 1.0% underwent VTE for the combined method. There was no statistical difference in VTE occurrence between the two methods. A graphical illustration of the results is shown in Figure 5.

4. Discussion

Venous thromboembolism (VTE) is a complex multifactorial clinical entity associated with significant morbidity and mortality and can present either as deep venous thrombosis (DVT) and/or concomitant pulmonary embolism (PE) [33]. VTE in malignancies is multifactorial-it can be cancer related as well as individual-related. A hypercoagulable state is induced directly by activating the blood clotting cascade and inducing pro-coagulant and inhibiting anticoagulant properties of endothelial cells, platelets, monocytes, and macrophages [34].

VTE does have a considerable impact on morbidity, mortality, and economic cost, and has led to the development of VTE risk stratification tools such as the Caprini risk assessment [33]. In addition to VTE risks associated with major surgical procedures and underlying malignancy, American Heart Association (AHA) has identified additional factors including prior VTE, age, obesity, immobility, and family history. Regarding the VTE risks associated with major surgical procedures and underlying malignancy, American Heart Association (AHA) has identified additional factors including prior VTE, age, obesity, immobility, and family history. VTE risk stratification in surgical patients according to the Seventh American College of Chest Physicians Conference on Antithrombotic and Thrombolytic Therapy is provided in Supplementary Table 4.

Thromboprophylaxis, whether mechanical, pharmacological, or combination is widely practiced across surgical specialties based on substantial evidence from randomized control trials (RCTs). However, there has been a lack of urology-specific evidence, leading to conflicting recommendations both nationally and internationally [35]. This gets more complicated when individual urological procedures are considered for VTE prophylaxis. In the case of RP, the European Association of Urology (EAU) 2022 has provided guidelines for VTE prophylaxis based on available literature [36]. The EAU has considered a few factors including patient VTE risk factors (Low/Medium/ High), type of surgical intervention (Open/laparoscopic/Robotic), Extent of surgery (No PLND/Standard PLND/Extended PLND), and availability of evidence for it. The summarized results are tabulated in Table.10. The studies and research used for EAU guidelines are mainly non-urological and rightly they have admitted that the evidence base is limited for this guideline. On the other hand, the National Institute for Health, and Care Excellence (NICE), UK, recommends VTE Prophylaxis to people undergoing abdominal (gastrointestinal, gynecological, urological) surgery who are at increased risk of VTE. It recommends mechanical VTE prophylaxis on admission for people undergoing abdominal surgery either anti-embolism stockings or intermittent pneumatic compression (until the patient is ambulatory). It also recommends Pharmacological VTE prophylaxis for people undergoing abdominal surgery, including-7 days for non-cancer and 28 days (Extended) for major cancer surgery postoperatively [37]. American urological society (AUA) recommends mechanical prophylaxis during MIS urological procedures, and pharmacological prophylaxis is reserved for high-risk patients only. For open urological procedures, AUA recommends the consideration of combined prophylaxis [38].

To our knowledge, this meta-analysis is the first of its kind specifically dealing with VTE prophylaxis in radical prostatectomy patients. We were able to establish a statistically significant higher incidence of VTE in open RP in comparison to MIS prostatectomy procedures. A similar observation has also been noticed in other abdominal procedures as well [39]. This analysis also revealed a higher incidence of VTE in patients undergoing RP as well as PLND. A study in 2011 including 773 patients, also demonstrated that there is a significant association between venous thromboembolism and radical prostatectomy plus pelvic lymph node dissection compared to radical prostatectomy only [40]. Lastly, we were able to establish that combined prophylaxis (Mechanical and pharmacological) does not have an advantage over mechanical prophylaxis in RP. This implies that PP is not indicated for the routine MIS approach for RP procedures. This is a finding which has not been confidently proven before this meta-analysis.

Additionally, we did find that there are other well-established factors that can contribute to VTE occurrences in post-RP patients as well, including age, personal and family history of VTE. Some of the studies in this meta-analysis have discussed these contributing factors, however, the data was not enough to perform a statistical analysis. These have been summarized in Table 3. These factors are essential in stratifying high-risk cohorts of patients.

Lastly, we found that the duration of VTE prophylaxis after RP has been quite variable across the globe. This is an issue that has been a bone of contention across surgical specialties both within and outside, and unfortunately, has been as such for more than a decade [41].

To summarize, this meta-analysis and the systematic review revealed that VTE events in RP procedures were reduced by switching to MIS procedures in comparison to open procedures. Additionally, PLND increases the risk of VTE and should be considered a high-risk factor while stratifying patients for consideration of VTE prophylaxis. Lastly, PP in cases of RP does not have an added advantage to mechanical prophylaxis. This systematic review forms a basis and provides a platform for future, randomized controlled trials taking into account all variables for consideration of globally acceptable guidelines for VTE prophylaxis in RP surgeries.

The limitations of this study include the scarcity of research available, the heterogenous nature of studies, and the non-uniformity of VTE prophylaxis practices across the globe for RP. In the future, there is a need for a multicentric randomized control trial to establish a VTE risk scoring system for RP patients and VTE prophylaxis guidelines to bring uniformity to this controversial practice.

5. Conclusion

This systematic review concludes that the overall incidence of VTE post-RP is 1%. The incidence of VTE in patients undergoing RP for prostate cancer is significantly less in minimally invasive (Laparoscopic/Robotic) procedures in comparison to open. Patients need to be stratified into risk groups using pre-existing guidelines and risk categories while PLND should be considered an additional risk factor. Additional PP to mechanical methods may not be necessary in all cases and should be considered in high-risk patients only.