Effects of proprioceptive neuromuscular facilitation combined with threshold inspiratory muscle training on respiratory function in neurocritical patients with weaning failure: a randomized controlled trial.

The same unblinded investigator gave all the treatments and did all the assessments, which the authors say may have introduced bias; maximum inspiratory pressure is a volitional test, so the person measuring matters. PNF was given for only five days, and because the trial was not powered for or designed around the training being superior to lung-protective ventilation, the authors say this may be too short to change expiratory volume in ventilated patients. No follow-up was done, so nothing is known about whether the change lasted or whether it affected weaning, time in ICU or survival. Patients had to be able to take part actively in training, and those with heart failure, neuropathy or myopathy, a tracheostomy before weaning, or a major neurological comorbidity were excluded, so this does not describe patients who cannot participate. Fifty patients were randomised and three were lost (one died, two transferred to another centre); the results table reports 24 control and 23 experimental patients. The results text says the experimental group had significantly greater diaphragm excursion than the control group, while the results table and the discussion say there was no effect on diaphragm excursion.

The authors declare there are no conflicts of interest. The text gives no separate funding statement. The trial was registered on the Chinese Clinical Trial Registry.

Do not read this as PNF improving breathing in general. The only measure where the two groups differed was maximum inspiratory pressure, by 4.37 cm H2O. Diaphragm movement, diaphragm thickening fraction and FEV1/FVC were unchanged, there was no follow-up, and on the oxygenation index the paper reports the control group improved more than the PNF group.