PDF: STEPS
(2016)Objective
To test whether pharyngeal electrical stimulation (PES) reduces radiological aspiration in patients with subacute poststroke dysphagia
Study Summary
• 58% of patients randomized to PES received suboptimal stimulation current (<10.2 mA), which may have contributed to the neutral result.
Intervention
Pharyngeal electrical stimulation (PES) via nasally inserted catheter with ring electrodes, delivered at 5 Hz for 10 minutes daily for 3 consecutive days, at 75% of the difference between sensory threshold and tolerance levels
Inclusion Criteria
Age ≥18, ischemic or hemorrhagic stroke within 42 days, clinical dysphagia confirmed by bedside screening (TOR-BSST), PAS ≥3 on videofluoroscopy
Study Design
Arms: Array
Patients per Arm: PES: 87, Sham: 75
Outcome
Bottom Line
Pharyngeal electrical stimulation was safe but did not improve aspiration at 2 weeks compared to sham in subacute stroke patients with dysphagia. 58% of patients randomized to PES received suboptimal stimulation current (<10.2 mA), which may have contributed to the neutral result.
Major Points
- Phase III trial following 3 positive pilot studies that showed PES reduced aspiration (PAS) and dysphagia severity (DSRS) in earlier-phase testing.
- 162 patients randomized (87 PES, 75 sham) across 20 sites in 5 countries (Denmark, France, Germany, Spain, UK). Mean age 74, 58% male, 89% ischemic stroke, mean PAS 4.8 at baseline.
- Primary outcome (PAS at 2 weeks): PES 3.7 (SD 2.0) vs Sham 3.6 (SD 1.9), adjusted difference 0.14 (95% CI -0.37 to 0.64, P=0.60) — no significant difference.
- Critical issue: 58% of PES patients received suboptimal stimulation (treatment current <10.2 mA, or identical threshold and treatment levels). Mean treatment current was only 14.5 mA with mean duration 9.8 minutes.
- No significant differences in any secondary outcomes: DSRS, mRS, Barthel Index, EQ-5D, PAS at 12 weeks, feeding route, or nutritional measures.
- Safety was reassuring: no serious adverse device-related events. Chest infection rates did not differ significantly (PES 21 vs Sham 11, P=0.19). All-cause mortality similar between groups.
- The undertreatment problem highlights the technical challenge of delivering adequate pharyngeal stimulation — the catheter electrode positioning and current titration protocol may need refinement for future trials.
Study Design
- Study Type
- Randomized, sham-controlled, patient-masked, outcome assessor-masked
- Randomization
- Yes
- Blinding
- Patient-masked and outcome assessor-masked (not treating researcher)
- Sample Size
- 162
- Follow-up
- 12 weeks
- Centers
- 20
- Countries
- Denmark, France, Germany, Spain, United Kingdom
Primary Outcome
Definition: Penetration Aspiration Scale (PAS) at 2 weeks
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 3.6 (SD 1.9) | 3.7 (SD 2.0) | - (-0.37 to 0.64) | 0.60 |
Limitations & Criticisms
- 58% of PES patients received suboptimal stimulation — a critical protocol adherence issue that likely underpowered the active treatment arm
- Primary analysis was changed before unblinding from mean of all swallows to mean of worst swallow per bolus — raises concerns about selective outcome reporting
- Relatively small sample (162 randomized, only 126 with primary outcome data) with substantial attrition
- Heterogeneous stroke population (ischemic and hemorrhagic, anterior and posterior circulation, wide range of severity)
- Median 11 days from stroke to randomization — may have missed the optimal therapeutic window for cortical reorganization
- Single-blind design (treating researcher unmasked) introduces potential performance bias
Citation
Bath PM, et al. Stroke. 2016;47:1562-1570.