A Meta-analysis of Standard Versus Ultrasound-Assisted Catheter-Directed Thrombolysis in the Management of Acute Pulmonary Embolism

Abstract

Background: Standard catheter-directed thrombolysis (SCDT) harnesses the therapeutic benefit of systemic thrombolytics while minimizing bleeding complications in patients presenting with pulmonary embolism (PE). Ultrasound-assisted catheter-directed thrombolysis (USAT) theoretically improves upon SCDT by disrupting fibrin and increasing the surface area exposed to thrombolytic agent. However, it is unclear if this translates into improved outcomes.

Methods: A systematic search of prior publications comparing SCDT and USAT in patients with intermediate or high-risk PE was conducted. Primary outcomes of interest were bleeding events, ICU and hospital length of stay. Secondary outcomes included changes in pulmonary artery systolic pressure (PASP), mean pulmonary artery pressure (mPAP), and right ventricle to left ventricle diameter (RV/LV) ratio. Studies that lacked comparison groups were excluded. Bias assessments were performed using the Cochrane tools for randomized and nonrandomized studies. Data was collated utilizing the Cochrane Review Manager software, and all analyses assumed random effects.

Results: Our search yielded 7 observational studies and 1 randomized control trial. The studies included a total of 543 patients who underwent either SCDT (n = 273) or USAT (n = 270) for intermediate or high-risk PE. The synthesized analysis showed no significant differences in bleeding between the groups. There were no differences in ICU or hospital lengths of stay, changes in PASP, or mPAP. Reductions in RV/LV ratio were greater with SCDT (mean difference, -0.16; 95% CI, -0.27 to -0.06; P = .003).

Conclusions: In comparison to SCDT, USAT did not result in improved clinical or hemodynamic outcomes in patients presenting with PE. Results were limited by heterogeneity among the included studies.

Keywords: catheters; meta-analysis; pulmonary embolism; thrombolysis; ultrasound.

Graphical abstract

Central Illustration

Flow diagram of systematic search. Process for the systematic literature search according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 statement.

Figure 1

Thrombolytic dose and infusion time. Forest plots demonstrate (A) thrombolytic dose and (B) infusion time for standard catheter-directed thrombolysis (SCDT) and ultrasound-assisted catheter-directed thrombolysis (USAT). The green square indicates the mean difference for each study calculated by subtracting the average thrombolytic dose (in milligrams) or average infusion time (in hours) in the USAT group from that of the SCDT group. Horizontal lines indicate the 95% CI. Data obtained from the randomized control trial are shown within the dotted line box. The combined MD and 95% CI for each analysis are represented by a black diamond. There was marked variability between the procedure details of each study, but the combined groups had no significant differences in mean thrombolytic dose or infusion time for SCDT and USAT.

Figure 2

Bleeding events. Forest plots comparing rates of (A) major and (B) minor bleeding events. The green square indicates the odds ratio for each study calculated by dividing the incidence of bleeding events in the standard catheter-directed thrombolysis (SCDT) group divided by the incidence of bleeding events in the ultrasound-assisted catheter-directed thrombolysis (USAT) group. Horizontal lines indicate the 95% CI. Data obtained from the randomized control trial are shown within the dotted line box. The combined odds ratio and 95% CI for each analysis are represented by a black diamond. There were no differences in rates of major or minor bleeding events among the studies that reported these outcomes.

Figure 3

ICU and hospital length of stay. Forest plots comparing differences in length of (A) intensive care unit (ICU) stay and (B) overall hospital stay. The green square indicates the mean difference for each study calculated by subtracting the average length of stay (in days) in the ultrasound-assisted catheter-directed thrombolysis (USAT) group from that of the standard catheter-directed thrombolysis (SCDT) group. Horizontal lines indicate the 95% CI. Data obtained from the randomized control trial are shown within the dotted line box. The combined mean difference and 95% CI for each analysis are represented by a black diamond. There were no differences in ICU or overall hospital length of stay between the groups.

Figure 4

Hemodynamic parameters. Forest plots comparing changes in hemodynamic outcomes with standard catheter-directed thrombolysis (SCDT) and ultrasound-assisted catheter-directed thrombolysis (USAT). The change in pre and postprocedure parameters was first calculated by subtracting the preprocedure from the postprocedure mean. The mean difference (green squares) was then calculated for each study by subtracting the change in the USAT group from the change in the SCDT group. The 95% CIs are represented by horizontal lines. Data obtained from the randomized control trial are shown within the dotted line box. The combined MD and 95% CI for each analysis are represented by a black diamond. There were no differences in the reduction in (A) pulmonary artery systolic pressure (PASP) or (B) mean pulmonary artery pressure (mPAP) achieved with either intervention. SCDT resulted in greater reductions in (C) right ventricle–to–left ventricle diameter (RV/LV) ratio than USAT.

Supplemental Figure 1A

PRISMA 2020 CHECKLIST. Page numbers where each checklist item may be found are listed.

Supplemental Figure 1B

PRISMA 2020 CHECKLIST. Page numbers where each checklist item may be found are listed.

Supplemental Figure 2

PRISMA 2020 ABSTRACT CHECKLIST. Items included in the abstract are marked with an “X”. Information about funding can be found at the end of the manuscript and information about protocol registration is included in the “Methods” section.

Supplemental Figure 3

RISK OF BIAS ASSESSMENT. Results from the risk of bias assessments are summarized above. Each column represents a different domain of bias. Overall risks of bias were moderate or serious for all included observational studies.

Supplemental Figure 4

FUNNEL PLOTS TO ASSESS PUBLICATION BIAS. Vertical dashed line represents the summary estimate of the treatment effect and outer dashed lines represent the 95% confidence limits around the summary treatment effect. For binary outcomes: effect summary estimate log risk ratio ± 1.96 × standard error of summary log risk ratio for binary outcomes. For continuous outcomes: effect summary estimate mean difference ± 1.96 × standard error of summary mean difference. ICU = intensive care unit; mPAP = mean pulmonary artery pressure; PASP = pulmonary artery systolic pressure; RV/LV = right ventricle to left ventricle diameter.

Supplemental Figure 5

SENSITIVITY ANALYSES. Forest plots comparing outcomes for (A) major bleeding events, (B) minor bleeding events, (C) ICU length of stay, (D) hospital length of stay, and (E) change in RV/LV ratio with data from Avgerinos et al. excluded. The green squares indicate OR for categorical variables and MD for quantitative variables. 95% CIs are represented by the horizontal lines. There were no differences in rate of major (A) or minor (B) bleeding events. ICU (C) and hospital (D) lengths of stay were similar between cohorts. Importantly, there was no difference between reductions in RV/LV ratio in this analysis (E). CI = confidence interval; MD = mean difference; RV/LV = right ventricle to left ventricle diameter; SCDT = standard catheter-directed thrombolysis; USAT = ultrasound-assisted catheter-directed thrombolysis.