To the Editor
We read the article by Grasegger et al 1 regarding the radar-based RECCO Rescue System with great interest, in particular because we faced a very similar situation this past winter. A 39-year-old woman was completely buried by an avalanche while skiing off-piste in the Swiss Alps. She was neither equipped with an avalanche transceiver nor a RECCO reflector. The ski patrollers, alerted by her companion, detected a signal after a 5-minute search using the RECCO Rescue System that was later assumed to come from the victim’s mobile phone. She was extricated from a depth of 80 cm after a 30-minute burial. She was uninjured, and a scarf covering both her nose and mouth probably prevented asphyxia from snow inhalation.
The advantages and limitations of the RECCO Rescue System are well described by Grasegger et al 1 in their article. However, we would like to add 2 further comments.
The first comment pertains to the analysis of the survival rate of completely buried avalanche victims located with the RECCO Rescue System. Grasegger et al 1 suggest that the vicinity of the accident sites to the patrolled ski areas may explain the high survival rate. Although vicinity certainly plays a role, we think that this apparently high survival rate should be interpreted cautiously. The reported cases represent a case series, without explicit selection criteria, and may be prone to a selection bias originating from 2 distinct sources. The first may be selection by the manufacturer of cases with a good outcome. The second, importantly, is the potential selection bias by those reporting cases to the manufacturer. A careful analysis of the case series seems to support this hypothesis. In the first category of victims found via the RECCO Rescue System within 1 hour of burial time, 9 of 11 (82%) were found not to have experienced cardiac arrest. This high survival rate is very unusual, as usually only one third of victims survive after a 30-minute burial time. 2 The second category of cases is represented by those found 1 or several days later using the RECCO Rescue System, all of whom died. Here again, the potential for a selection bias is high, as such dramatic cases are also certainly prone to being reported.
The utility of the RECCO Rescue System in both these situations is undeniable, either to save life in specific conditions or to help find the fully buried corpse of a victim. However, the magnitude of its impact on survival or retrieval rates can only be determined if the true numerator (ie, the total number of survivors or retrieved corpses located using the RECCO Rescue System) is not inflated while the true denominator (ie, the total number of fully buried avalanches victims wearing the device) is not underestimated.
Grasegger et al 1 must have reached the same conclusion because, in the abstract of their report, they state that the publication of their case is intended to encourage the collection of cases in which fully buried people are found using the RECCO Rescue System. However, this intention is not further discussed in the article, and no proposition made to support this intention. We suggest 2 potential approaches. First, the manufacturer could promote the reporting of all cases found wearing the RECCO Rescue System, dead or alive, by their impressive network of ski patrollers. Alternatively, the corresponding author, who is affiliated with a recognized mountain medicine research organization, could also collect all cases through a dedicated web site, his personal e-mail address, or a dedicated e-mail address. This independent approach would also prevent any conflict of interest.
Both solutions would be an incentive for rescue teams to report cases in which the RECCO Rescue System was critical in finding a fully buried person, such as the case we were involved with this past winter.