Is a temperature difference of >2.2OC an accurate marker to detect future diabetic foot ulceration?

Since the 1950s, when Dr Paul Brand observed that feet “heat up before they break down” into ulceration, we have known that temperature may be a marker for detecting impending foot ulceration. In the 2000s, a number of large trials from Profs Larry Lavery and David Armstrong proved that home-monitoring of increasing temperature was very effective in detecting and then preventing impending foot ulcers. They used a difference of >4^O Fahrenheit (equivalent to >2.2^O Celsius) between feet as the threshold to detect impending ulceration based on their experience using a Temptouch device.

Yet, even with this good evidence how many people use home-monitoring of temperature in diabetic foot care? Not many we’d suggested. Why? Well it’s thought that the time required by patients to test their temperatures multiple times per day at home and the high numbers of false positive (detections) using this threshold turns patients and clinicians off using this method.

However, we recently reported on a large study from DFA’s friend Prof Bijan Najafi and colleagues that used a temperature sensing mat.  The study required patients to simply stand on the mat for 20 seconds each day at home and it automatically recorded temperature differences. The mat was found to be very user-friendly as most patients continued to use it for >6 months. Plus, they found the >2.2^O C threshold correctly identified 97% of diabetic foot ulcers a month in advance; but also had a high 57% false-positive rate, indicating an ulcer was coming when it didn’t. They then also tried a threshold of >3.2^OC and found this decreased the false-positive rate to 32%, but also correct identification to 70%.

This all means we now know that detecting temperature differences between feet does predict ulcers, we have easy ways for our patients to do this at home, but we still don’t have a temperature threshold that is perhaps accurate enough to detect ulcers without also giving too many false detections. Now a new study – by DFA’s Dr Jaap van Netten and Anke Wijlens published in the International Wound Journal – has investigated various definitions of this threshold.

The authors recruited 20 patients with neuropathy and no history of foot ulcers. They got the patients to measure the temperature of both of their feet in 6 locations, 4 times per day, for a week using a Temptouch. The locations were under the hallux, 1^st, 3^rd, 5^th metatarsal heads, midfoot and heel; and the times were when they woke up, lunch time, dinner time and before going to bed. They also cleverly monitored patients’ daily steps, the environmental temperature surrounding the patients, and the time of day the foot temperatures were measured. Last, they asked patients to contact them if they detected a >2.2^O C difference between feet at the same location and followed them for a week to see if they developed any ulcers. What they found was interesting.

First, they found one foot stayed warmer than the other throughout; in some patients it was their right foot and in others their left. Second, they found on average a 0.65^OC ‘natural’ difference between feet throughout the week; the Hallux had 1.1^OC difference and all other locations were around 0.6^OC. Third, they found that activity, environmental temperature and time of day had no impact on these temperature differences. Fourth, unfortunately (or fortunately) no patients developed ulcers so they couldn’t calculate sensitivity. But they could calculate false positives, because they knew any time patients recorded temperatures above the threshold it did not detect an ulcer coming.

They investigated three different threshold definitions to find out the % false-positives from all measurements taken and how many patients reported at least one false-positive in the week. They then adjusted these same three definitions for each patient’s individual ‘natural’ difference between feet to create an “individual threshold”. For example if the patient’s right foot was on average always 0.5% warmer than the left, then that patients adjusted individual threshold to exceed would be >2.7^OC on the right and >1.7^OC on the left foot to account for this natural temperature difference. The differences all had to be for the same location on the foot. They found for the threshold definition:

1. >2.2^OC on one occasion: 8.5% of all were false-positives & 95% patients reported >1 per week.
2. >2.2^OC on two consecutive occasions: 5% & 70%.
3. >2.2^OC same time on two consecutive days: 3% & 35%
4. >individual threshold on one occasion: 8% & 80%.
5. >individual threshold on two consecutive occasions: 9% & 45%.
6. >individual threshold same time on two consecutive days: 2% & 20%

The authors concluded, “The >2.2^OC threshold for impending ulceration is not valid for home monitoring to prevent foot ulceration when used as single measurement. The validity improves to acceptable levels with subsequent confirmation of an above-threshold recording the following day and further improves with individual correction.”  And the authors recommend “baseline temperature differences between the left and right foot need to be measured and taken into account for home monitoring of skin temperature, to optimise validity and reduce false-positive outcomes”.

As for any study this study had strengths and limitations. The strengths included: investigating a group of very similar neuropathic patients, measuring foot temperature multiple times each day for a week, also measuring activity, room temperature and time measured, and adjusting for individual natural differences in temperature. The limitations included: small sample size, not measuring severity of neuropathy or patients with a history of ulceration which may impact on temperature, and only following patients for one week to detect if they developed ulcers (not >1 month as Prof Najafi found).

Regardless, this study brings us a step closer to determining an accurate individualized temperature threshold for our patients to predict impending diabetic foot ulcers with enough time to prevent them. Larger longer studies are now required to see if these individualized thresholds do accurately detect ulcers with very low false positives …. and just quietly we believe these studies may be underway.

If such an individualised threshold proves to be accurate, then combining it with a temperature mat may be the simple “fire alarm” that saves your patients feet in future; or our reliable home “foot alarm”.