Two patient warming systems will be compared with 40 patients each in a convective warming group (BairHugger, Arizant) and in a conductive warming group (HotDog, Augustine Biomedical).

The patients will undergo small to medium trauma surgery and will be warmed with the randomized device - the hypothesis is, that the area under the standardized core temperature / time curve is significantly greater in the conductive warming group. Secondary outcome is the mean skin temperature / time area under the curve.

General study design We will study 200 patients (18-90 years) undergoing elective orthopedic surgery at the trauma surgery unit. The patients must have normal weight (20-30 BMI), the duration of surgery should last between 3 - 4 hours. There will be no other exclusion criteria (except severe peripheral arterial disease in the warmed extremity), as forced air patient warming is routinely used for all patients during this procedure.

The patients will be randomly assigned via a computer generated randomization list to treatment with either polymer or forced air warming. Anaesthesia and fluid management will be administered as desired by the anaesthesiologist.

Before warming four skin temperature probes will be attached to the upper arm, chest, abdomen and back of the patients to calculate mean body temperature. A probe (Mallinckrodt Anesthesiology Products, Inc., St. Louis, MO) will be introduced in the ear to measure core temperature. Afterwards, treatment with BairHugger or HotDog will be started.

The measurements will be recorded every five minutes until the end of surgery when intraoperative warming is stopped.

Differences between core temperatures will be analyzed both by using the summary measure of AUC (area under the curve) and by comparing the core-temperature at the end of surgery in the polymer and the forced air group with two-tailed, unpaired t tests if normally distributed and with a Wilcoxon test, if not normally distributed (distribution tested by K-S-test).

Results are expressed as means ± standard deviations if normally distributed, as median (IQ-Range) when not normally distributed.

Differences will be considered statistically significant when P < 0.05. We consider a difference in core temperature at the end of surgery of 0.5°C as clinically important. Considering this difference as well as a known standard deviation of approximately 1.5°C we will have to study 40 patients in each arm of the study to achieve statistical significance using a power of 90% and a p-value of 0.05.

Inclusion Criteria:

• The patients must have normal weight (20-30 BMI), the duration of surgery should last between 3 - 4 hours.
• There will be no other exclusion criteria , as forced air patient warming is routinely used for all patients during this procedure.

Exclusion Criteria:

• Severe peripheral artery disease in the warmed extremity