Phrenic nerve amplitude as an effort-independent prognostic biomarker of respiratory failure in amyotrophic lateral sclerosis.

OBJECTIVE: In amyotrophic lateral sclerosis (ALS), the effort-dependent respiratory metrics used to determine the timing of non-invasive ventilation (NIV) lose reliability as bulbar and/or cognitive impairment progresses. We tested whether the effort-independent phrenic nerve compound muscle action potential (CMAP) amplitude predicts the timing of NIV independently of forced vital capacity (FVC) and the revised ALS Functional Rating Scale (ALSFRS-R). METHODS: Pre-registered analysis of a single-centre ALS cohort (n = 1,486) contributing 2968 serial phrenic nerve conduction studies. Amplitude was expressed as percentage of an age- and sex-predicted value (per 10% decrease), and time to NIV initiation was the primary outcome (death competing). We used cause-specific Cox models, time-dependent and bootstrap-validated discrimination, and a Bayesian joint longitudinal-survival model. RESULTS: Lower amplitude predicted earlier NIV (adjusted hazard ratio [HR] 1.22 per 10% decrease, 95% CI 1.16-1.27; likelihood-ratio χ2 = 77.6, p ≈ 10⁻1⁸). The association was concentrated in the first year (HR 1.31, 1.24-1.38) and attenuated thereafter. The optimism-corrected discrimination increment over the clinical model was +0.055 (0.034-0.073); below-normal amplitude approximately doubled 12-month NIV risk (62% vs 29%). In the joint model the current amplitude (HR 1.57, 1.46-1.69) and its rate of decline (HR 1.30, 1.20-1.44) were independently prognostic. CONCLUSIONS: Phrenic nerve conduction studies, an effort-independent neurophysiological marker carried robust, near-term, incremental prognostic information for NIV beyond FVC and ALSFRS-R, establishing diaphragmatic motor amplitude as a candidate biomarker for objective respiratory prognostication in ALS. External validation and recalibration are required before clinical use.