The intradermal capsaicin models have also been used to show increased hyperalgesic reactions in patients with rheumatoid arthritis (Morris et al. 1997), fibromyalgia (Morris et al. 1998), and vulvodynia (Foster et al. 2005).

The capsaicin model has been applied to non-painful conditions such as patients with multiple chemical sensitivity and showed facilitated hyperalgesic reactions (Holst et al. 2011).

Transdermal electrical stimulation is another way to induce central sensitization, and larger hyperalgesic areas have been shown in patients with chronic lumbosacral pain including radicular neuropathic features as compared to controls (Lecybyl et al. 2010).

One study has used the capsaicin model in patients for pharmacological profiling. Patients with unilateral sciatica showed heightened responses to intradermal capsaicin compared to pain-free volunteers, and the study by Sumracki et al. (2012) compared the effects of pregabalin (300 mg) and the tetracycline antibiotic and glial attenuator minocycline (400 mg) on capsaicin-induced spontaneous pain, flare, and allodynia and between the affected and unaffected leg. It was concluded that a healthy control group was needed for such studies in order to show differences as a pain patient cannot be used as his/her own control.

Temporal summation and aftersensation (Fig. 2 ): An important and potent central mechanism in dorsal horn neurons is the temporal integration also termed wind-up-like pain. The initial phase of the wind-up process observed in animals translates into temporal summation in humans (Arendt-Nielsen et al. 1994). If a painful stimulus is repeated 1–3 times per second, the pain will integrate and become more painful with an intensity and stimulus frequency-dependent summation (Arendt-Nielsen et al. 1994). Temporal summation can be elicited using electrical, mechanical, or thermal stimulation modalities and is elicited from the skin, musculoskeletal structures, and viscera (Arendt-Nielsen 1997; Arendt-Nielsen and Yarnitsky 2009).

In many clinical conditions when repeated stimuli are applied in neuropathic, musculoskeletal, and visceral chronic pain patients, temporal summation is substantially facilitated (Arendt-Nielsen et al. 1997a; Graven-Nielsen et al. 2000; Nikolajsen et al. 1996).

In clinical bed site testing, simple devices are used for assessing temporal summation such as tapping the skin with a nylon filament (Nikolajsen et al. 1996). When more standardization is required, however, automated user-independent methods are needed such as thermal (Kong et al. 2013), mechanical (Nie et al. 2009), or electrical stimulation techniques of the skin (Arendt-Nielsen et al. 1994), muscles (Arendt-Nielsen et al. 1997b), or viscera (Drewes et al. 1999). The abovementioned cuff-algometry technique can also be used if the cuff is repeatedly inflated (Lemming et al. 2012; Skou et al. 2013).

When repeated stimuli are delivered, sometimes pain patients experience an aftersensation (pain after the stimulus has stopped) (Robinson et al. 2010). This has been observed in patient with neuropathic (Gottrup et al. 2003) and musculoskeletal pain (Staud et al. 2003, 2007). An exclusive facilitation of the aftersensation alone has been proposed to be of diagnostic value (Sato et al. 2012) and supports the basic finding that the summation and the aftersensation are mediated by different underlying central mechanisms (Price et al. 1978; You et al. 2005).

Facilitated temporal summation in chronic pain patients is efficiently inhibited by NMDA receptor antagonists (ketamine and amantadine). This has been found in patients with surgical incisions (Stubhaug et al. 1997), postherpetic neuralgia (Eide et al. 1994), phantom limb pain (Nikolajsen et al. 1996), chronic postsurgical neuropathic pain (Pud et al. 1998), and fibromyalgia (Graven-Nielsen et al. 2000).

Abnormal temporal summation in patients with neuropathic pain does not predict the clinical effect of imipramine or gabapentin (Rasmussen et al. 2004).

Some clinical studies have not shown effect on facilitated temporal summation by lamotrigine (an antiepileptic drug acting on voltage-sensitive sodium channels) in patients with spinal cord injury pain (Finnerup et al. 2002) and by the NMDA-antagonist memantine in patients with phantom limb pain (Nikolajsen et al. 2000) and postherpetic neuralgia (Eisenberg et al. 1998).

More clinical studies are needed to understand which compounds can modulate temporal summation.

Spatial summation (Fig. 3 ): Spatial summation is another mechanism, which relies on central networks and the general sensitization status (Bouhassira et al. 1995). In humans, spatial summation can be assessed in different ways where the stimulus is applied to using different stimulation areas, e.g., using thermodes (Nielsen and Arendt-Nielsen 1997), pressure probes (Nie et al. 2009), or cuff algometry (Polianskis et al. 2002). The cuff-algometry technology can utilize one or two cuffs which are automatically inflated, and the volunteer/patient rates the provoked pain intensity (Polianskis et al. 2002). Spatial summation is facilitated in various pain conditions such as fibromyalgia (Staud et al. 2004, 2007), osteoarthritis (Graven-Nielsen et al. 2012), and lateral epicondylitis (Jespersen et al. 2013).

No pharmacological screening studies have used this technique in pain patients.

Reflex receptive fields (Fig. 4 ): In many animal studies, expansion of receptive fields of dorsal horn neurons has been documented in neuropathic as well as inflammatory (cutaneous, muscle, and viscera) models. Expansion of receptive fields has been a challenging mechanism to assess in humans. In animals, an alternative method for assessment of receptive field expansion of dorsal horn neurons has been developed. This involves quantification of the so-called reflex receptive field (Schouenborg et al. 1994, 1995) which was found to expand in the presence of central sensitization (Harris and Clarke 2003).

This method based on assessment of the nociceptive withdrawal method has been translated to humans (Andersen et al. 2001), and the reflex receptive field is found to be expanded into spinal cord injured patients (Andersen et al. 2004), in patients with chronic visceral pain (endometriosis) (Neziri et al. 2010), in patients with low back pain (Biurrun Manresa et al. 2013), and in patients with neck pain (Biurrun Manresa et al. 2013).

To date, few pharmacological studies have used this technique for profiling, and the sensitivity is not known. One study showed that the 5-HT-3 receptor antagonist had no effect on the expanded fields in chronic low patients and no effect on the clinical pain (Neziri et al. 2012). This advanced technique may in the future provide new information in clinical studies and for profiling of new compounds.

Descending pain modulation (Fig. 5 ): There are increasing evidences that the balance between the descending pain inhibition and facilitation may be disturbed in chronic pain and that this phenomenon has a role in maintaining central sensitization (Porreca et al. 2002; Suzuki et al. 2004; You et al. 2010). In humans, status assessment of the descending pathways has recently undergone a revival, and the original DNIC terminology has been renamed to conditioning pain modulation (CPM, Yarnitsky et al. 2010). Less efficient CPM seems to be a general phenomenon in chronic pain conditions (Staud 2012) such as patients with, e.g., myofascial temporomandibular joint pain (Bragdon et al. 2002), chronic low back pain (Peters et al. 1992), fibromyalgia (Kosek and Hansson 1997), whiplash (Daenen et al. 2014), painful OA (Arendt-Nielsen et al. 2010), and chronic tension-type headaches (Sandrini et al. 2006), chronic pancreatitis, and interstitial cystitis (Ness et al. 2014). OA patients with a deficient, descending pain inhibition show a normalization to a pain-free state after surgery (Graven-Nielsen et al. 2012; Kosek and Ordeberg 2000) suggesting that the chronic pain maintained the CPM dysfunction and that the chronic pain saturated the CPM mechanism so the conditioning pain stimulus is less efficient.

The CPM paradigm has in recent years been implemented in pharmacological screening studies, and modulatory effects can be induced by dexmedetomidine (a selective α(2)-adrenoceptor agonist) (Baba et al. 2012), buprenorphine (Arendt-Nielsen et al. 2012a, b), fentanyl (Arendt-Nielsen et al. 2012a, b), duloxetine (Yarnitsky et al. 2012), and apomorphine (a non-specific dopamine agonist) (Treister et al. 2013), whereas oxycodone seems not to modulate CPM (Suzan et al. 2013).

The use of CPM for assessing drug efficacy in pain patients has only been used in one study so far where the effect of duloxetine in painful diabetic neuropathy was investigated (Yarnitsky et al. 2012).

Evidently, an alteration in the descending pain modulation could be a promising target for pharmacological intervention (Arendt-Nielsen and Yarnitsky 2009) and should be utilized when new pain drugs are evaluated.

Offset analgesia (Fig . 6 ): Offset analgesia is provoked by incremental decreases of a nociceptive heat stimulus, and the perception of the provoked pain is less as what is provoked when the same stimulus intensity is given as a single stand-alone stimulus (Grill and Coghill 2002; Yelle et al. 2008). A peripheral component of offset analgesia cannot be excluded but is generally considered as a central inhibitory modulation of pain and as such should be modulated by central sensitization. This is supported by the fact that an association between CPM and offset analgesia has been shown indicating some commonalities of their underlying mechanisms (Honigman et al. 2013).

In healthy volunteers, offset analgesia is found stable in one study (Nilsson et al. 2013) but in another study found to be facilitated when assessed in repeated sessions. Furthermore, the response is reduced in potency with age (Naugle et al. 2013). Induction of thermal hyperalgesia has shown an association with increased magnitude of offset analgesia (Martucci et al. 2012a, b), whereas other studies have tried to modulate the potency of offset analgesia by experimental central sensitization without success (Martucci et al. 2012a, b), suggesting that longer-lasting central sensitization is needed.

Only one clinical study has used the offset analgesia paradigm in patients with neuropathic pain and found the response was reduced or absent (Niesters et al. 2011a, b). More clinical studies are needed to confirm the usefulness of this technique.

The offset analgesia is not modulated by intravenous ketamine (Niesters et al. 2011a, b) or morphine (Martucci et al. 2012a, b). More studies are needed to understand the role of this technique in pharmacological screening.

Referred pain (Fig. 7 ): Referred pain is a central phenomenon predominantly related to pain from deep somatic and visceral structures. Referred pain areas are facilitated as a result of a central sensitization in patients with, e.g., fibromyalgia (Sorensen et al. 1998), whiplash (Koelbaek Johansen et al. 1999), osteoarthritis (Arendt-Nielsen and Yarnitsky 2009), chronic low back pain (O’Neill et al. 2007), irritable bowel syndrome (Swarbrick et al. 1980), and chronic pancreatitis (Dimcevski et al. 2007). In addition, it has been shown that referred pain areas evoked from muscles are enlarged in visceral pain conditions (Bajaj et al. 2003) as a result of convergence and interaction between sensitization related to deep somatic and visceral structures but also manifested as viscero-visceral sensitization (Giamberardino et al. 2010). Patients with, e.g., urinary calculosis show significant and long-lasting hyperalgesia in the referred pain areas (Giamberardino et al. 1994); patients suffering from a large number of colics show more hyperalgesia than those experiencing a limited number of colics. Similar somatic hyperalgesia is found in referred pain areas in acute appendicitis (Stawowy et al. 2002) and cholecystolithiasis (Stawowy et al. 2005).

Few studies have used the referred pain areas in pain patients for pharmacological profiling. Ketamine (Arendt-Nielsen and Graven-Nielsen 2008), gabapentin (Arendt-Nielsen et al. 2007), and a 5-HT antagonist (Christidis et al. 2008) are found to reduce the area of referred pain. On the contrary, alfentanil and morphine do not reduce referred pain (Schulte et al. 2006).