No strong evidence about the effects of sensory stimulation for a brain-injured person in a coma

About half of people in a coma (deep unconsciousness) because of traumatic brain injury will wake within a year of the accident. Speeding recovery to allow people to wake sooner is a priority for them and their family. One type of treatment uses sensory stimulation to try to keep the person's brain working normally. Sensory stimulation methods vary greatly, from one or two hourly sessions of a day, through to shorter sessions every hour for 12 to 14 hours a day. The review found there is no strong evidence to determine whether sensory stimulation benefits people in comas.

Authors' conclusions: 

This systematic review indicates that there is no reliable evidence to support, or rule out, the effectiveness of multisensory programmes in patients in coma or vegetative state.

Read the full abstract...

Coma and vegetative state follow traumatic brain injury in about one out of eight patients, and in patients with non traumatic injury the prognosis is worse. The use of sensory stimulation for coma and vegetative state has gained popularity during the 1980's but beliefs and opinions about its effectiveness vary substantially among health professionals.


To assess the effectiveness of sensory stimulation programmes in patients in coma or vegetative state.

Search strategy: 

We searched the Injuries Group specialised register, the Cochrane Controlled trials register, EMBASE, MEDLINE, CINAHL and PSYCHLIT from 1966 to January 2002, without language restriction. Reference lists of articles were scanned and we contacted experts in the area to find other relevant studies.

Selection criteria: 

Randomised or controlled trials that compared sensory stimulation programmes with standard rehabilitation in patients in coma or vegetative state.

Data collection and analysis: 

Abstracts and papers found were screened by one reviewer. Three reviewers independently identified relevant studies, extracted data and assessed study quality resolving disagreement by consensus.

Main results: 

Three studies were identified with 68 patients in total. The overall methodological quality was poor and studies differed widely in terms of outcomes measures, study design and conduct. We therefore did not carry out any quantitative synthesis but reviewed results of available studies qualitatively.