A systematic review of contemporary randomized controlled trials for treatment of lower extremity lymphedema by...

Observational trials have suggested that intermittent pneumatic compression (IPC) devices reduce lymphedema (LED) volume and improve patient-reported outcomes, but this type of trial is subject to bias. Systematic reviews (SRs) of randomized controlled trials (RCTs) provide the highest evidentiary strength for examining the impact of a therapeutic intervention and are important for evidence-based clinical practice guidelines. Few existing SRs, however, include recent RCTs and only a smaller number address LED of the lower extremities (LELED). Therefore, we conducted a SR based on PICO questions to identify RCTs that examined the effect of IPC on LELED for a clinical practice guideline.

For this SR, English-language RCTs in CINAHL, Embase, and MEDLINE published between January 1, 2010, and January 31, 2025, were searched and the PRISMA reporting method was used. Both intragroup and intergroup comparisons of the specified outcome measures were examined for statistical significance. The evidentiary quality of the RCTs and risk of bias were assessed using GRADE and the Cochrane risk of bias tool, respectively.

In the PRISMA analysis, 141 articles were identified through a search of the databases and an additional 4 articles were discovered through other sources, for a total of 145 eligible RCTs. Twenty-one RCTs addressed upper extremity LED and seven LELED RCTs met the criteria for inclusion: adults ≥18 years with unilateral or bilateral chronic LELED and present for ≥6 months and in the maintenance phase of treatment. The patient must have an ankle-brachial index of ≥0.8, have no active infection or acute venous thrombosis, and be able to use the study device and comply with daily therapy. The average total number of patients per RCT was 54, and the mean patient age was 55 years. Female gender predominated (67%). The majority of RCTs were composed of secondary LED, but three series included a segment of primary LED. Chronic venous insufficiency was the most common etiology in the seven RCTs. Five RCTs had two arms, and two had three arms. The experimental arm used an advanced pneumatic compression device in five RCTs, but complex decongestive therapy was combined in this arm with IPC in three. The control comparator arm was complex decongestive therapy in three, a simple IPC in one, and an advanced pneumatic compression device using a different mode in another. Circumferential measurement of limb girth by tape was the most common method to determine volume changes. Five of seven RCTs used a daily treatment regimen, with four using twice daily sessions. Study length varied with 4 weeks in three, 5 weeks in one, and 8 months in another. An a priori power calculation was carried out in three of the seven RCTs. Reduced limb volume was the outcome measure used in all seven studies and all seven showed statistically significant reductions in limb volume when comparing baseline values with end-of-treatment values. Five RCTs also demonstrated differences in their intergroup comparisons of edema reduction. Patient-reported outcomes in five RCTs with three generic quality-of-life (Short Form-36, visual analog scale, and EQ5D3L) and two disease-specific (Lymph ICF-LL and LYMQOL) quality-of-life measures demonstrated significant improvements by intragroup comparison in three RCTs. Heterogeneity and the composition of the experimental and control groups, however, prevented meta-analysis. The methodological quality of the seven RCTs included in this review was evaluated using the Cochrane Risk of Bias Tool. Overall, the included studies demonstrated moderate methodological quality. Assessment of these RCTs by GRADE showed moderate certainty that IPC reduces lower limb edema, volume, or circumference compared with standard care or no IPC.

Despite combining (bundling) additional treatment with IPC in three studies, intragroup comparisons demonstrated highly significant reductions in limb volumes with IPC in all RCTs and in five of seven RCTs by intergroup comparison. Quality indicators demonstrated a moderate risk of bias and moderate certainty.