Reproducibility and influence of test modality order on thermal perception and thermal pain thresholds in quantitative sensory testing
Introduction
Clinical assessment of the functional integrity of myelinated peripheral nerve fibres is routinely done by a combination of electroneurograhy (ENeG), which provide indirect quantitative estimates of the degree of axonal and demyelinating damage, and electromyograhy (EMG), which provide additional estimates of possible axonal damage in myelinated motor fibres (Kimura, 2001). However, ENeG techniques are not feasible for evaluation of the function in thin myelinated or unmyelinated nerve fibres, as their electrophysiological contribution to the electrically induced compound nerve impulse volley will be proportionally small and also further decreased by temporal dispersion and phase cancellations between heterogeneous types of fibres. Some aspects of the functional integrity of thin fibres may be evaluated by direct intra-neural microneurograhic recordings (Gandevia and Hales, 1997, Vallbo et al., 2004), but such methods are not fully adopted for clinical routine work. Further, development of techniques for evaluation of somatosensory evoked potentials by painful laser stimulation of the skin also bring additional means for thin fiber evaluation, but these estimates of impulse propagation velocity then also include the function of the ascending non-specific somatosensory tracts (Truini et al., 2004, Cruccu et al., 2008). Thus, at present there are few electrophysiological methods available for direct clinical assessment of peripheral small nerve-fiber function.
Alternatively, the integrity of thin myelinated and unmyelinated fibres may be indirectly assessed by semi-quantitative psychophysical testing of detection thresholds for temperature and temperature pain, usually denoted ‘quantitative sensory testing’ (QST) (Gruener and Dyck, 1994, Yarnitsky, 1997). Several rationales for QST have been described, which all have in common that the skin is warmed or cooled with a Peltier element, while the patient non-verbally reports the perception of a thermal sensation by pressing a switch (Fruhstorfer et al., 1976, Gruener and Dyck, 1994, Gelber et al., 1995, Dyck and O’Brien, 1999). The recorded thermal thresholds are taken as indirect indicators of the functional state of the somatosensory pathways supplying the tested area. However, it should be kept in mind that such thresholds have a multifactorial basis, as the thermal percept depends not only on intact peripheral innervation, but also on intact ascending somatosensory pathways, central processing, and descending motor pathways and effector organs (Kandel et al., 2000). Also technical details may influence the results, such as the initial and the adapted skin temperature under the stimulating probe, the size of the probe, or the rate of temperature change during stimulations, as well as the patient’s reaction time and degree of mental concentration (Pertovaara and Kojo, 1985, Yarnitsky and Ochoa, 1991, Hilz et al., 1995a, Hilz et al., 1995b, Chong and Cros, 2004). With all these precautions kept in mind, QST is a widely accepted method for thin fiber evaluation in clinical (Yarnitsky and Ochoa, 1991, Hilz et al., 1995a, Hilz et al., 1995b, Shy et al., 2003, Cornblath, 2004, Hanson et al., 2007) as well as research settings (Shy et al., 2003, Hanson et al., 2007). However, no testing strategy for thermal QST have yet been accepted as the single method of choice, and consequently reported estimates of threshold reference values and reproducibility of the assessments vary with the different methods (Jamal et al., 1985, Armstrong et al., 1991, Yarnitsky and Ochoa, 1991, Gruener and Dyck, 1994, Yarnitsky, 1997).
In clinical settings, an often adopted strategy for QST is the ‘method of limits’ (MLI) (Fruhstorfer et al., 1976, Hilz et al., 1999), which by comparison is rather quick and easy to perform, and which enables descriptions of the thermal detection capacity in an easily intelligible way. However, in previous studies this method has shown a varying degree of reproducibility and variability (Levy et al., 1989, Yarnitsky and Sprecher, 1994, Hilz et al., 1995a, Hilz et al., 1995b, Yarnitsky et al., 1995, Becser et al., 1998, Fillingim et al., 1999, Meier et al., 2001, Moravcová et al., 2005, Palmer and Martin, 2005, Lowenstein et al., 2008, Wasner and Brock, 2008). Unfortunately variations in methodological strategies and statistical analyses makes direct comparisons between studies difficult (Chong and Cros, 2004), but even with a highly standardized rationale, a high variability in the actual threshold assessments would reduce the usefulness of the test by confounding individual test data and obscuring reference materials. Thus, to evaluate the usefulness of the MLI it is crucial to try to confirm the earlier observations of temperature threshold assessment variability.
The purpose of the present study was to evaluate the reproducibility of thermal perception and thermal pain thresholds, assessed with QST by the MLI in a group of healthy young volunteers. The aim was to describe the intra- and inter-individual variability and reproducibility of thermal cold and warmth perception thresholds, as well as of cold-pain and heat-pain perception thresholds, both for longer testing intervals between days and for shorter intervals within the same day.
Section snippets
Subjects
Thirty-eight healthy volunteers (18 males and 20 females, aged 22–55 years; mean 30.9) participated in the study. Twenty-nine subjects were tested on three occasions at 3 days (days 1, 2 and 7), and, in an additional reduced series, nine subjects were tested at only one occasion each day, likewise at three different days. Subjects received both verbal and written information about the experimental procedure. They had all given their informed consent according to the World Medical Association’s
Results
When reviewing the results of the statistical comparisons, a general observation was that all calculations of putative differences between samples, as well as calculations of coefficients of repeatability (CR), showed comparable results whether based on relative or absolute thermal threshold estimates. In the following, only the computations based on relative thresholds are presented, except for the evaluation of coefficients of variability (CV).
Discussion
The main finding of the present study was that thermal and thermal pain thresholds in QST with the ‘method-of limits’ (MLI) demonstrate a high degree of reproducibility. Furthermore, it was also shown that the perception of cold and warm was influenced by the sequential order in which different modalities were tested, in that preceding determination of thermal pain thresholds influenced thermal perception threshold assessments.
Conclusion
Differences in both study methodologies and statistical approaches contribute to the difficulties and confusion in comparisons of QST reproducibility reports. In the present study, selected rationales from earlier reports were adopted to re-evaluate of the reproducibility of the “method-of-limits” over shorter and longer time intervals. We show that quantitative sensory testing with this method is a reliable rationale for routine clinical use, and can be considered as a feasible tool for
Acknowledgements
Mr. Magnus Edlund for skilful help with data conditioning. Parts of the present material were collected and compiled as a part of a Bachelor Degree in the Biomedical Sciences at Umeå University for one of the authors (L.S.). The study was supported by grants from the Research foundation of the Department of Pharmacology and Clinical Neuroscience, and by the J.C. Kempe Memorial Found for Scholarships (SJCKMS scholarship), both Umeå University, and by grants from the patients’ associations FAMY
References (45)
- et al.
Recommendations for the clinical use of somatosensory-evoked potentials
Clin Neurophysiol
(2008) - et al.
The methodology and scope of human microneurography
J Neurosci Methods
(1997) - et al.
Components of variance for vibratory and thermal thresholds testing in normal and diabetic subjects
J Diabetes Complications
(1995) - et al.
Thermal perception thresholds: influence of determination paradigm and reference temperature
J Neurol Sci
(1995) - et al.
Influence of the rate of temperature change on thermal thresholds in man
Exp Neurol
(1985) - et al.
Quantitative sensory testing in the German research network on neuropathic pain (DFNS): standard protocol and reference values
Pain
(2006) - et al.
Laser evoked potentials for assessing sensory neuropathy in human patients
Neurosci Lett
(2004) - et al.
Determinants of thermal pain thresholds in normal subjects
Clin Neurophysiol
(2008) - et al.
Thermal testing: normative data and repeatability for various test algorithms
J Neurol Sci
(1994) - et al.
Heat pain thresholds: normative data and repeatability
Pain
(1995)
A study of peripheral diabetic neuropathy. The applications of age-related reference values
Diabet Med
Normal distributions of thermal and vibration sensory thresholds
Muscle Nerve
Reliability of cephatic thermal thresholds in healthy subjects
Cephalalgia
Thermal threshold testing for the assessment of small fibre dysfunction: normal values and reproducibility
Diabet Med
Methods of measurements of thermal thresholds
Acta Neurol Scand
Technology literature review: quantitative sensory testing
Muscle Nerve
Diabetic neuropathy: diagnostic methods
Adv Stud Med
Quantitative sensation testing in epidemiological and therapeutic studies of peripheral neuropathy
Muscle Nerve
Sex differences in heat pain thresholds as a function of assessment method and rate of rise
Somatosens Mot Res
Method for quantitative estimation of thermal thresholds in patients
J Neurol Neurosurg Psychiatry
Cited by (73)
Interoceptive technologies for psychiatric interventions: From diagnosis to clinical applications
2024, Neuroscience and Biobehavioral ReviewsCorneal Confocal Microscopy Abnormalities in Children and Adolescents With Type 1 Diabetes
2023, Endocrine PracticeA novel thermoelectric device integrated with a psychophysical paradigm to study pain processing in human subjects
2023, Journal of Neuroscience MethodsNew directions in research on somatic concerns in individuals with eating disorders
2022, New Ideas in PsychologyCitation Excerpt :Though not as commonly used as heartbeat counting tasks, there are several other objective measurements of interoception, including thermosensation tasks (Heldestad, Linder, Sellersjö, & Nordh, 2010), taste tasks (Frank et al., 2003), and odor tasks (Hummel, Sekinger, Wolf, Pauli, & Kobal, 1997). Thermosensation tasks manipulate temperature (Heldestad et al., 2010), and while sensing temperature is an important interoceptive process, manipulation of temperature does not elicit strong emotions (Crucianelli, Enmalm, & Ehrsson, 2021). Though this task may be particularly useful in studies designed to account for affective states, the task itself may not elicit emotions and cognitions associated with ED psychopathology, such as guilt and feeling fat (Fairburn, 2008), thus making it challenging to detect associations between interoception and affect and how that relates to eating disorder symptoms.
- 1
Present address: The SDS-Clinic, Vasaplatsen 8, SE-411 34 Gothenburg, Sweden.