Methods

Eight experienced Russian cold water swimmers, aged 29–53 years, participated in a 92 km relay swim in Finnish lakes starting 29 May 2000 wearing one piece bathing suits (women) or trunks (men). All participants swam once daily, on five or six days of the six day event. We recorded data for the longest of three to five swims by each when water temperature was <11°C. Each swimmer planned in advance the approximate duration of each swim with the relay team organiser (VL) but then varied this time on the basis of sensations during the swim. After each swim, the swimmer rewarmed in a sauna on the attendant launch. Height and weight were measured for each swimmer. Skinfolds at four standard sites—over the biceps, lower corner of the scapula, costal margin below the midpoint of the clavicle, and the abdomen 50 mm below and lateral to the umbilicus—were measured with Harpenden calipers. Table 1 presents the mean of the skinfold measures for each swimmer. Rectal temperature was measured at 120 mm, and recorded in resting conditions in the cabin of the launch in air at 18–25°C before swims and at its coldest point in the afterdrop 6–25 minutes after the swims.

Results

Mean skinfold thickness varied between 8.56 and 21.5 mm (table 1). Surface area/mass ratio of the swimmers varied only between 0.022 and 0.027 m²/kg and is not reported in detail.

Table 1 shows that people with less thick subcutaneous fat made significantly shorter swims than those with thicker fat layers (p = 0.035). The thinnest subject swam for only 23 minutes, whereas the four with the thickest fat layers swam for over an hour. The leanest swimmer had the largest fall in rectal temperature, 2.9°C, and the lowest rectal temperature after the swim, 34.3°C, but, overall, the falls were not significantly related to fat thickness.

Discussion

The data imply that the thinner swimmers largely compensated for faster heat loss by shorter swims, and were able to judge when it was necessary to leave the water as their core temperature fell to near an unacceptable level. Three swimmers cooled below 35°C, the accepted upper limit of hypothermia, but even the lowest core temperature recorded, 34.3°C, represents no threat to life in people able to rewarm in quiet surroundings. Self selection among these experienced cold water swimmers is likely to account for the absence of any very thin people, who would be subject to rapid cooling.

The results do imply risks if highly competitive conditions had caused them to prolong the swims. Body temperatures below 32°C⁶ cause a progressive risk of spontaneous ventricular fibrillation and death, and the coldest swimmer would have reached that level within about a further 20 minutes if the rate of cooling had continued. Colder water must also increase the risk, both because of faster body cooling and because pain followed by anaesthesia of the skin in very cold water could disturb ability to assess body cooling. In the 3.5°C water of Lake Baikal, a man who swam for 25 minutes had a rectal temperature of 32.2°C on leaving the water, and rewarmed without apparent harm (N Lando, unpublished work). Monitoring of body core temperature by radiotelemetry² may be needed throughout very competitive endurance swims and in any endurance swims in water colder than 9°C. This applies particularly to swimmers who lack extensive cold water experience backed up by measurements of fat thickness and rectal temperature.

CONCLUSIONS

Experienced cold water swimmers under experienced supervision and without competitive pressure could safely make prolonged swims in water at 9–11°C, judging the duration of their swims only by subjective sensation.