Assessment of Somatosensory Function and Self-harm in Adolescents

This cross-sectional study assesses somatosensory function in adolescents and compares their somatosensory profiles with community control participants to investigate associations with the incidence and frequency of self-harm.


eAppendix. Description of quantitative sensory tests (QST) used in the study.
The DFNS QST protocol measures cutaneous sensitivity to a range of stimulus modalities including temperature, light touch, strong pressure and vibration. The tests provide a complete sensory profile for one region of skin within 30 minutes while the participant is seated in a comfortable chair. The purpose is to provide insight into peripheral-and central pain and touch processing mechanisms via subjective reporting of stimulus detection and pain thresholds. The protocol is highly standardised allowing data to be compared across different study populations. The DFNS has published skin-region specific, normative reference values from a large cohort of healthy controls stratified by age and sex. Individual data are z-transformed using reference values to determine whether individual data are within the range of normal variability. For full details on the reference values, including range of possible scores in their original units, and z-transformation methods we refer the interested reader to Magerl et al., (2010)  Thermal detection and pain thresholds were tested using the computerised TSA-II NeuroSensory Analyser (Medoc, Israel). Water was circulated through a 3.2 cm2 thermode (a Peltier device) secured against the forearm by a Velcro strap. Temperature change was programmed to occur from a baseline temperature of 32 °C, with heating and cooling delivered at a rate of 1 °C/sec to a maximum of 50 °C and 0 °C, respectively (interstimulus interval: 6 sec; baseline return rate: 8 °C/sec). Each threshold test was performed three times. For assessment of cold and warm detection thresholds (CDT and WDT), participants were asked to press a stop button as soon as they noticed a change in temperature above or below baseline temperature. The arithmetic means for both CDT and WDT were calculated using the difference between the measured threshold temperature and the baseline temperature (32 °C). Following, the thermal sensory limen (TSL) was assessed which is a detection threshold test using alternating warming and cooling temperatures (six temperature changes in total) to assess paradoxical heat sensations (PHS). Participants pushed a stop button when they noticed a change in temperature from baseline and reported the temperature sensation (e.g., 'warm'/'cold'/'burning'). The TSL was calculated by subtracting the arithmetic mean of the CDT (three measurements) from that of the WDT (three measurements). For assessment of cold and heat pain thresholds (CPT and HPT), participants were asked to press a stop button at their first pain sensation during warming and cooling of the thermode. The arithmetic mean of three repetitions was calculated using the absolute temperature values (°C) for CPT and HPT.
The mechanical detection threshold (MDT) was assessed using a set of standardised von Frey hairs (a pen-like instrument with a nylon thread attached; Optihair, Marstock Nervtest, Germany) ranging from 0.25 -256 mN. Using a modified method of limits protocol, the hairs were lightly touched against the skin with increasing and decreasing intensity, and the participant was asked to indicate whether they detected the touch sensation. The test was performed in an 'up-down' pattern until five supra-and five subthreshold responses were taken. The geometric mean was calculated as the MDT. The mechanical pain threshold (MPT) was performed using weighted pinprick stimuli (MRC Systems GmbH, Germany) ranging from 8 -512 mN. The pinprick tips have a flat contact area of 0.25mm in diameter. Like MDT, the test was performed in an 'up-down' pattern until five supra-and five subthreshold responses were taken. The geometric mean was calculated as the MPT.
Mechanical pain sensitivity (MPS) and dynamic mechanical allodynia (DMA) (stimulus-response function model) were assessed using a combination of light-touch stimuli (cotton wool, a soft brush and a cotton bud) and pinprick stimulators touched against the skin in a pseudorandom order. Participants were asked to rate the painfulness of each stimulation using a numeral rating scale anchored at 0 (no pain) and 100 (worst pain imaginable). A total of 50 stimulations were given and pain sensitivity was calculated by the geometric mean of the pain ratings given for pinprick stimuli (MPS) and innocuous stimuli (DMA) separately.
Wind-up ratio (WUR) was assessed using a 256 mN pinprick stimulator. Participants were asked to rate the painfulness of a single stimulation (numerical rating scale 0-100). This was followed shortly after by a short train of 10 stimuli (repetition rate of 1/sec). At the end of the train, participants were asked to provide a second pain rating. This was repeated five times in a (non-overlapping) small region of skin. For each of the five test repetitions, the arithmetic mean ratio was calculated by dividing the pain rating for the stimulus train with that of the single stimulus. The geometric mean was calculated from these ratios to generate a WUR score. If a rating of '0' was given three times for the single stimulus (i.e., three of the five repetitions) the test was not continued as WUR could not be calculated.
Vibration detection threshold (VDT) was assessed using an activated 64 Hz tuning fork (Ragg, Sheffield, UK) which was placed on the bony prominence at the wrist. The VDT was determined from the "wandering" tip of a triangle indicated along a scale on the fork and moved by means of the vibration. Participants were asked to indicate the point at which they no longer perceived the vibration sensation. The test was performed three times and the arithmetic mean was calculated.
Pressure pain threshold (PPT) was assessed using a manual pressure algometer (Wagner Instruments, USA) pressed into the muscle at the base of the thumb. Pressure was increased at 0.5 kg/sec and participants were asked to say stop once the pressure first became painful. The test was repeated three times and the arithmetic mean (in kPa) was calculated as threshold.