ORIGINAL ARTICLE

Association Between a Composite Score of Pain Sensitivity and Clinical Parameters in Low-back Pain Søren O’Neill, DC, PhD,*wz Claus Manniche, MD, DMSc,wz Thomas Graven-Nielsen, DMSc,* and Lars Arendt-Nielsen, DMSc*

Background and Aims: A limited number of quantitative sensory pain tests (QST) were selected on the basis of ease of application and interpretation in a clinical setting. QST results were summarized as a composite score on a scale of 0 to 4 that was deemed to facilitate clinical interpretation. The QST set was used to investigate differences in pain sensitivity between low-back pain (LBP) subgroups and was correlated with important clinical parameters. Materials and Methods: A total of 198 LBP patients and 44 controls were recruited and data were collected on: LBP intensity and duration, improvement in LBP, functional status, psychological profile, quality of life, and current work situation. QST consisted of pressure-pain threshold of the infraspinatus muscle, pain intensity with sustained mechanical pressure on the thumb nailbed, conditioned pain modulation, and pain tolerance assessed by coldpressor test. Results: Chronic, but not acute LBP patients scored significantly higher on the composite score of pain sensitivity than controls. A weak linear correlation between the composite score of pain sensitivity and intensity of clinical LBP and disability was found. A contingency was found between a high pain sensitivity (dichotomized composite score of pain sensitivity) and the lack of improvement in LBP. Pain sensitivity did not correlate with LBP duration, psychological profile, or sick leave. Conclusions: Pain sensitivity may be important for the prognosis of LBP, but QST is not currently part of routine clinical examination of LBP patients. The selected set of pain tests and the composite score of pain sensitivity could serve as a clinically applicable QST procedure in the examination of LBP. Key Words: low-back pain, quantitative sensory testing, generalized hyperalgesia, central sensitization, experimental pain

(Clin J Pain 2014;30:831–838)

Q

uantitative sensory testing (QST) is the psychophysical testing of a test participant’s response to a controlled stimulus of known intensity. Typical examples include the pressure-pain threshold (PPT), that is, the indication of onset of pain when exposed to gradually increasing Received for publication March 2, 2013; revised November 29, 2013; accepted September 30, 2013. From the *Laboratory for Human Experimental Pain Research, Department of Health Science and Technology, Center for SensoryMotor Interaction, University of Aalborg, Aalborg; wSpine Center of Southern Denmark, Lillebælt Hospital, Middelfart; and zClinical Locomotion Science Network, University of Southern Denmark, Odense, Denmark. The authors declare no conflict of interest. Reprints: Søren O’Neill, DC, PhD, Rygcenter Syddanmark, Medicinsk Rygafdeling, Sygehus Lillebælt, Middelfart Sygehus, Østre Hougvej 55, Middelfart DK-5500, Denmark (e-mail: soeren.oneill@ rsyd.dk). Copyright r 2013 by Lippincott Williams & Wilkins DOI: 10.1097/AJP.0000000000000042

Clin J Pain



Volume 30, Number 10, October 2014

mechanical pressure. Unlike clinical pain, the intensity of nociceptive stimulus is controlled in QST and in pain research, QST has been used with success to study differences and changes in pain perception and processing, both in the laboratory and in clinical settings. In relation to low-back pain (LBP), QST has been used to demonstrate generalized hyperalgesia in chronic LBP patients compared with pain-free controls.1–5 Conversely, no difference in pain sensitivity has been found between controls and acute experimental or subacute clinical LBP.4,6 Furthermore, there is QST evidence that experimental pain sensitivity correlates with important clinical parameters, such as physical function and psychological distress3,7 in LBP patients. QST may potentially provide important clinical information in the assessment of painful conditions. Efforts have been made to establish standards for QST of neuropathic pain.8 Conversely, QST is not part of a typical clinical evaluation of patients with LBP and important discriminatory differences may exist between diagnostic groups,9,10 which suggests that perhaps different QST paradigms are appropriate for different clinical conditions. Previous QST studies on LBP patients1–5,7,11 have employed a heterogeneous variety of QST procedures in different combinations, including mechanical pressure, electrical stimulation, injection of hypertonic saline, pinprick, and others. Most of the studies, however, have been on highly selected subgroups of LBP patients: primarily chronic patients seen in hospital settings. If QST is to be used effectively in the assessment of LBP, a standard set of QST procedures needs to be developed which is employable in busy clinical practice and which provides an adequate assessment of pain sensitivity. It is not immediately obvious, however, how best to adequately assess pain sensitivity. Previous research has demonstrated poor correlation between different QST measures of pain sensitivity.12,13 More complex statistical procedures such as component-analysis, factor-analysis, and clusteranalysis suggests, that although some QST procedures reflect the same underlying component, others do not.14–17 Most of these studies have been conducted on pain-free volunteers and none has included LBP patients. The findings suggest, however, that several different pain measures are necessary to gain an adequate picture of pain sensitivity. In particular, QST procedures should be selected which employ different pain modalities (mechanical, thermal, electrical, etc.) and different response domains (detection threshold, tolerance, response intensity measures, etc.). In the present study, a candidate for a small standard set of QST tests to assess generalized hyperalgesia in LBP was constructed and tested in a heterogeneous group of LBP patients and controls. We hypothesized that the set of QST tests would (1) demonstrate generalized hyperalgesia www.clinicalpain.com |

831

Clin J Pain

O’Neill et al

in chronic, but not acute LBP patients compared with pain-free controls; (2) correlate with clinical LBP, improvement in LBP, physical function, psychological parameters, and quality of life; and (3) be feasible for clinical use.

MATERIALS AND METHODS Participants As LBP is a heterogeneous condition, participants were recruited from 3 different populations to obtain a broad sample: hospital outpatients, primary care patients, and pain-free volunteers.

Hospital Outpatients The Spine Center of Southern Denmark is the largest spine center in Denmark; a publicly funded, regional outpatient hospital department with a medical and surgical section attending to some 15.000 new patients every year. Patients with long-lasting spinal pain syndromes, who have failed to improve sufficiently with treatment in the primary health care setting, are referred from general medical practitioners, chiropractors, medical consultants, and other hospital departments. Over a period of 2 months, new patients from the outpatient clinic of the Spine Center of Southern Denmark were invited to participate in the study. Participants were recruited consecutively on predesignated (pseudorandom) days. Patients who agreed to participate were interviewed and examined (QST as described below) immediately before or after a prescheduled magnetic resonance imaging scan.



Volume 30, Number 10, October 2014

Pressure-Pain Threshold Pressure was applied manually with an algometer (Somedic, Ho¨rby, Sweden; 1 cm2 probe) on the left infraspinatus muscle, with a near-constant velocity of 50 kPa/s until the participant indicated the pressure as becoming painful by pressing a button connected to the algometer. PPT measurements were repeated 3 times with 5- to 10second interval and the average was used for further analysis. The algometer was placed such that repeat applications overlapped partly. If no pain had been elicited by 1000 kPa, this was recorded as the PPT. If the first and second measurement was 1000 kPa, a third was not performed.

Clamp-sustained Pressure A spring clamp (Millarco 100 mm with two 14  13 mm2 pressure pads, Prod.Nr. 72644; Millarco, Lystrup, Denmark) exerting a pressure of approximately 5 kg (1.8 cm2) was used (Fig. 1). The spring clamp was placed on the left thumbnail for 10 seconds, after which participants were asked to score the pain intensity on a 100mm-long visual analogue scale (VAS) on paper, marked “No Pain” and “Worst Pain Imaginable” at either end. Care was taken to ensure that the upper pad of the clamp was placed as far proximal on the nail as possible but not overlapping the eponychium (cuticula) and the lower pad proximal enough to prevent the clamp sliding forward. A few participants had a particularly short or curved nail, and it was necessary to manually prevent the clamp from sliding forward. The same clamp was used for all patients, for both spring clamp tests (before and after cold-pressor test [CPT]).

Primary Care Patients Patients were recruited from 2 primary care chiropractic clinics. Over a period of 7 predesignated days, new patients and patients scheduled for follow-up visits where invited to participate, if their primary symptom was LBP (with/without other musculoskeletal symptoms). Potential participants were identified by the attending clinician in connection with the clinical consultation. Only those deemed eligible for inclusion by the clinician were referred to and invited to participate by the principal researcher (S.O.N.). Thus, no data were collected on patients who declined to participate or who were deemed by the clinician to be ineligible for inclusion. Following QST, the same clinical information as for hospital patients was collected.

Pain-free Volunteers

Cold-Pressor Test A 25 L water-tub (Dometic Waeco Mobicool C40, Dubai, United Arab Emirates) was kept refrigerated at 0 to 21C and the temperature was confirmed by means of a thermometer immediately before each CPT. Participants were instructed to submerge their right, nonclenched hand to the wrist in the circulating water (Reich submersible waterpump, 15 L/min, 0.5 bar; Reich, Arnhem, Netherlands) and keep it there for 1 minute, and not to withdraw their hand, unless the pain became unbearable. Time from submersion to withdrawal was recorded when participants could not tolerate the full 1 minute CPT. Participants were informed when 30 and 50 seconds had passed. Time from submersion to withdrawal was recorded and participants were categorized as CPT tolerant

Pain-free volunteers were recruited and examined by QST. No clinical data were collected, apart from verbal confirmation that the participant did not report any pain symptoms.

Ethics The experiment was conducted in accordance with the Helsinki Declaration and IASP guidelines for human pain research.18 Approval was granted by The Regional Scientific Ethical Committee for Southern Denmark (ref. S-20090003).

Experimental Protocol The participants were positioned comfortably, seated for the QST examination and data were collected by S.O.N. or by one of 2 research assistants.

832 | www.clinicalpain.com

FIGURE 1. Spring clamp. Generic spring clamp, exerting a pressure of approximately 5 kg (1.8 cm2). r

2013 Lippincott Williams & Wilkins

Clin J Pain

Volume 30, Number 10, October 2014

Composite Score of Pain Sensitivity and Clinical Parameters in LBP

or nontolerant, depending on their tolerance of the full 60 seconds of CPT.

patients in primary care and were completed 15 to 30 minutes following QST. Patients were asked to report the typical number of days per week with LBP, the approximate date of LBP onset (from which LBP duration was calculated), and to indicate whether their LBP was “improved” or “no change or worse,” within the last 2 weeks, and since debut. LBP duration of >3 months (91 d) was considered chronic. The psychological profile screening questionnaire consisted of 8 questions with answer options on a 7-point Likert scale. The questions were partly based on the PRIME-MD and Orebro Musculoskeletal pain questionnaires (Table 1). A validation study on the screening questionnaire, which concludes that “The concurrent validity of these screening questions was comparable to, or better than, alternate questions previously reported,” has been accepted for publication.24 Work situation was assessed by the following questions: “Have you been sick-listed because of LBP in the last 3 months?”; “If so, for how many days?”; “Are you still sick-listed?”; “Are you on reduced work hours because of your LBP?”; and “What is your current work situation?” (answer options: regular job, flex-job, student, in rehabilitation, unemployed, early retirement, retired, home-maker, other). For the purposes of the current study, regular employment was defined as either regular job, student, retired, or home-maker.



Conditioned Pain Modulation (CPM) Sustained mechanical pressure on the left thumbnail with the spring clamp (Fig. 1), as described above, was repeated 5 to 15 seconds after CPT, thus serving as teststimulus in assessment of CPM, with CPT as conditioning stimulus. On the basis of the change in spring clamp pain response before/after CPT, participants was categorized as inhibitory CPM responders (reduced pain response by 5 mm VAS or more), facilitatory CPM responders (increased pain response by 5 mm VAS or more) or CPM nonresponders (a difference in pain response of