To the Editor:

We appreciate Mr. Zusman's comment on the possible missing link between impaired descending pain control systems and widespread hypersensitivity detected at sites away from the painful knee in osteoarthritis patients. The terminology used in the area of descending inhibition is somewhat confusing, since the terms “DNIC,” “DNIC-like effects,” and “counterirritation” are all used in the literature. It was therefore recently suggested that the term “conditioned pain modulation” should be adopted (1). An additional rationale for the name change was to avoid a direct comparison between the human manifestation (i.e., changed pain sensitivity during conditioning pain stimulus) and the very specific neurophysiologic mechanisms demonstrated in animal studies on descending inhibition and facilitation.

In animal studies, a number of neuronal structures (e.g., periaqueductal gray matter, rostral ventromedial medulla, medullary reticularis nucleus dorsalis, and subnucleus reticularis dorsalis) are found to be included in descending inhibition and facilitation where some of the centers driving descending inhibition are opioid sensitive. Under normal conditions there is a delicate balance between descending inhibition and facilitation, where inhibition is normally greater than excitation (2, 3). After tissue injury, alterations in this balance result in a loss of inhibition and/or an increase in excitation (4). These basic findings support clinical data indicating that the balance between inhibition and facilitation is altered in chronic pain states, and that the descending pain modulatory pathways could be a key component in development of chronic pain. It has been suggested that the maintenance of central sensitization in animal models of neuropathy requires descending pain facilitation mechanisms driven from the rostral ventromedial medulla (5). One important finding in animals, as well as in humans, is that the descending modulation affects the entire neuroaxis, with the best effect occurring extrasegmentally to the conditioning stimulation (6). Hence, inhibition or facilitation may cause generalized widespread hypoalgesia or hyperalgesia, respectively.

Mr. Zusman suggests that an impaired descending inhibitory control system in clinical pain conditions is not likely to explain the widespread hypersensitivity found in our study or other studies of musculoskeletal pain patients. The reasoning is that pain sensitivity outside the painful knee will be unaffected when the descending inhibitory systems are impaired; however, in that case the facilitatory system might be boosted. In human experimental and clinical studies we can measure only the net sum of inhibition and facilitation and hence cannot decide which of the systems are affected in, for example, chronic musculoskeletal pain conditions, such as fibromyalgia or painful knee osteoarthritis. Interestingly, in our study, we also found that pressure pain sensitivity in osteoarthritis patients was further increased on the knee when conditioning pain stimulus was administered. This further supports the notion that there is a change in the balance between inhibitory and facilitatory descending control.

One important factor is the relevance of descending control in chronic pain. Functional imaging studies of patients with hip osteoarthritis show abnormal activity in the brainstem periaqueductal gray matter, an area known to be involved in the modulation of mechanical stimuli via descending modulatory pathways (7). The extent to which chronic pain will activate the descending inhibitory control system has been debated. One interpretation is that the descending inhibition is fully activated or saturated by the pain, and as such cannot cause any further inhibition during the conditioned pain modulation. Even though the descending system is working at its full capacity, the effect may be cancelled by the concurrently activated facilitatory system. In such cases the manifestation of widespread hypersensitivity may be observed. In addition, the ongoing localized nociceptive activity (from the osteoarthritic knee, for example) may cause localized hyperalgesia. Animal studies have shown that unilateral pain may cause bilateral sensitization, documenting the contralateral spread of sensitization (8). There are also experimental and clinical data suggesting that the spreading sensitization depends on the pain intensity and duration (9–11).

Overall, findings in animal studies document the descending inhibitory and facilitatory mechanisms acting in a balance under normal conditions, although they may be disturbed after tissue injury. At present, the balanced effects can be assessed in humans, but the mechanisms cannot be differentiated into the inhibitory and facilitatory components. In our view, there are no data that indicate the need to revise the current concept of the role of descending control in mechanisms of widespread hypersensitivity and chronic pain.

T. Graven-Nielsen DMSc, PhD*, T. Wodehouse PhD†, R. M. Langford MBBS, FRCA, FFPMRCA†, L. Arendt-Nielsen DMSc, PhD‡, B. L. Kidd MBChB, DM, FRCP§, * Aalborg University, Aalborg, Denmark, † Barts and The London School of Medicine and Dentistry, London, UK, ‡ Aalborg University, Aalborg, Denmark, § Barts and The London School of Medicine and Dentistry, London, UK.