IPC for prevention of upper limb venous thrombosis with PICCs

30 January 2026 Daniel Stubbins Intensive Care Medicine, Summaries Leave a comment

Intermittent pneumatic compression can reduce the incidence of upper extremity venous thrombosis after peripherally inserted central catheter placement in traumatic brain injury patients: A randomised control trial

Liang et al. Australian Critical Care 2025; 38(6). doi:10.1016/j.aucc.2025.101308

Clinical Question

In patients with traumatic brain injury (TBI) on neurointensive care with peripherally-inserted central catheters (PICC), does the use of upper-limb intermittent pneumatic compression (IPC) reduce the rate of upper-extremity venous thrombosis (UEVT) compared to standard care?

Background

PICCs are commonly inserted in the intensive care setting as a means of long-lasting central venous access
Their presence is associated with an increased rate of venous thrombosis, especially in the catheterised limb
TBI patients are known to be at further increased risk of venous thromboembolism, potentially amplifying this risk
Pharmacological thromboprophylaxis may be higher-risk in some TBI patients due to the ongoing risk of intracerebral haemorrhage
Usage of lower-limb IPC as a DVT reduction strategy in immobile patients with stroke was studied in the 2013 CLOTS 3 trial. This found significantly lower DVT rates in the IPC group
A 2020 study looking at the use of upper-limb intermittent pneumatic compression for TBI patients with PICCs was terminated early due to a signal of increased DVT in the intervention group, although there was a high rate of crossover with a low sample size making interpretation difficult

Design

Single-blinded parallel-group RCT
- Data collectors and assessors blinded
Randomisation: 1:1 allocation to control or intervention group through computer-generated sequence printed in opaque sequential envelopes
Power calculation: Assuming a baseline UEVT rate of 38%, and the intervention reducing this to 20%, 96 patients would be required in each arm to achieve over 80% power at the 5% significance level. Increased to 120 per arm to account for a 20% dropout rate

Setting

Neurointensive care unit in a single tertiary hospital in China (Shanghai General Hospital)
Patients were recruited from January 2021 to December 2023

Population

Inclusion:
- Patients with TBI and no autonomous movement with GCS 8-10
- Underwent PICC insertion
- Normal platelets (125-350 x 10⁹/L), prothrombin time (9-13s) and D-dimer (<0.5 mg/L)
- No personal history of venous thrombosis or cancer
- No swelling, redness, heat or pain indicative of phlebitis in the upper limbs prior to PICC insertion
- Able to comply with vascular ultrasound examination
- Family consent
Exclusion:
- PICC insertion below the elbow
- Unsuccessful first-time PICC insertion
- Automatic discharge or death within the 28-day IPC treatment period
285 patients screened. 240 randomised. 195 included for analysis (98 intervention, 97 control)
- Of those excluded at screening: 31 not meeting criteria, 14 declined to participate
- Of those removed prior to analysis: 18 for failed first-time PICC insertion, 1 for insertion site below the elbow, 18 self-discharged (9 in each group) and 8 died (3 intervention, 5 control)
Baseline characteristics of intervention vs control group:
- Age: median 62 vs 61
- Sex: 45.9% vs 50.5% male
- Cardiovascular history: 6.1% vs 4.1%
- Diabetes history: 13.3% vs 15.5%
- PICC laterality: 81.6% vs 83.5% right (protocol dictated right basilic vein as first-line site)
- Catheter depth: median 38cm vs 38cm
- Arm circumference: median 26cm vs 26cm
- Post-hoc analysis found no significant difference between the groups in any of these characteristics

Intervention

The use of an upper-limb IPC system on the cannulated arm
- Applied to the forearm with a cycling pressure gradient for 20 mins every 8 hours (approx. 1 hour daily per patient)
- Initiated 24 hours after PICC insertion
- 28-day treatment period
- If UEVT identified, ICP treatment was stopped

Control

‘Standard care’
- Not further defined, but without IPC devices

Management common to both groups

Ultrasound-guided placement of a 4-French single-lumen PICC
A standard PICC care bundle, including dressing changes, pulsatile saline flushing and daily arm circumference measurement
If UEVT identified then the patient was assessed for risk of intracerebral haemorrhage and given subcutaneous low-molecular weight heparin if low-risk
Scanned at 7, 14 and 28 days with additional scans triggered if >0.5cm increase in arm circumference
Followed-up for 2 months

Outcome

Primary outcome: Incidence of UEVT in the catheterised arm
- Defined as presence of hypoechoic changes within the venous lumen on ultrasound, reduced or absent colour flow in the lumen on doppler ultrasound, or decreased venous compressibility on increasing probe pressure
- Significantly lower in the intervention group (4.1% vs 18.6%, p = 0.001)
Secondary outcomes: condition of deep (axillary, subclavian and brachiocephalic) and superficial (cephalic and basilic) veins, maximum blood flow velocity in the basilic vein (measured at 0, 7, 14 and 28 days), and incidence of additional complications (superficial phlebitis and pulmonary embolism)
- Significantly lower in intervention group:
  - Deep vein thrombosis (1% vs 8.2%, p = 0.018)
  - Superficial vein thrombosis (3.1% vs 10.3%, p = 0.042)
- Significantly greater in intervention group:
  - Basilic vein peak velocity at 14 days (median 10.1 vs 9.0m/s, p < 0.001)
  - Basilic vein peak velocity at 28 days (median 10.6 vs 8.9m/s, p < 0.001)
- No significant difference in:
  - Basilic vein peak velocity at 0 days (median 10.8 vs 10.8m/s, p = 0.978)
  - Basilic vein peak velocity at 7 days (median 9.8 vs 9.5m/s, p = 0.054)
- Superficial phlebitis: 1% vs 6.2% (no statistical analysis reported)
- Pulmonary embolism: none reported in either group

Authors’ Conclusions

IPC treatment can reduce the incidence of UEVT in patients with TBI post PICC placement and increase upper limb vein blood flow

Strengths

Achieved target numbers of patients in each analysis group as per power calculations
Well-matched groups on characteristics reported
Uniform catheter selection (4 Fr, single lumen) with no difference in median insertion depth – a known risk-factor for UEVT

Weaknesses

Unclear definition of ‘standard care’, with no mention of numbers receiving pharmacological thromboprophylaxis or lower-extremity IPC in the group characteristics, or listing either as exclusion criteria
Many patient risk factors for PICC-associated venous thrombosis were not reported, including BMI, smoking status, mechanical ventilation or any surgical intervention undertaken
All velocity measurements were taken from the catheterised basilic vein only, with no flow analysis on deeper vessels, making any link between this and more significant DVTs unclear
Excluding patients who died within the study timeframe from analysis may have excluded any potentially relevant deaths related to venous thromboembolism (although none listed by the authors)
The duration of treatment (3x 20 mins sessions/day) was much lower than is generally applied by lower-limb IPC protocols and on which much of the existing evidence is based

The Bottom Line

This RCT found a significant reduction in the incidence of UEVT in the IPC group, contrary to previous research which has found increased rates
There are substantial limitations around lack of clarity relating to control group care and certain relevant patient characteristics
With these in mind, this study alone will not change my practice, but it justifies further studies into mechanical prevention of PICC-associated venous thrombosis

External Links

Metadata

Summary author: Daniel Stubbins
Summary date: 21^st December 2025
Peer-review editor: David Slessor

Picture by: S Daboul/Pexels