CYCLOOXYGENASES (COX) and prostaglandins are key players in inflammatory diseases and contribute significantly to the accompanying pain sensitization. More than 10 yr of research have shown that, in particular, those prostaglandins that are produced by the inducible COX-2 isoenzyme trigger inflammatory reactions in the tissue. Two articles in this issue of the Journal now suggest that the constitutively expressed COX-1 might be similarly important for the development of neuropathic pain—at least in animal models. Zhu and Eisenach 1show that spinal COX-1 expression increases early in experimental neuropathy. Hefferan et al.  2provide data suggesting that inhibition of COX-1 during early stages prevents the development of two typical symptoms of painful neuropathies: allodynia, which describes a state of increased pain sensation in response to stimuli that are usually not sensed as painful, such as light touch; and hyperalgesia, which is an increased sensitivity to noxious (painful) stimuli.

Both studies were conducted in closely related standard animal models of neuropathic pain. Zhu and Eisenach 1used the partial peripheral nerve transsection 3and Hefferan et al.  2the peripheral nerve ligation model. 4Because both of these models involve surgical procedures, they are associated with tissue damage and trigger some inflammatory response. They are therefore not universally accepted as “good models” resembling the most frequent forms of neuropathic pain in patients, which occur in the course of metabolic diseases such as diabetes or renal failure. 5Nevertheless, both groups have performed reasonable controls to show that the inflammatory component was, at least, not dominating.

If we assume that the results of both groups can be transferred to the clinical situation of patients, e.g. , after traumatic nerve injury, their results bear important consequences for the treatment or prevention of neuropathic pain. Unlike inflammatory pain, neuropathic pain is difficult to treat. Classic cyclooxygenase inhibitors as well as opioids are only marginally effective, and physicians often use anticonvulsants and drugs with unknown mechanisms of action, such as gabapentin, with variable success. The present studies may provide a rational basis for an early, or possibly even prophylactic, treatment of neuropathic pain. In light of the short time period, such a prophylactic intervention will not be possible in metabolic neuropathies. However, the present results may promote clinical studies in patients with acute nerve injuries. One might speculate that cyclooxygenase inhibitors might be given as premedication before elective surgery when the patient is at risk for the development of painful neuropathies (e.g. , before amputation). Cyclooxygenase inhibitors might therefore find a place in so-called preemptive analgesia in neurosurgery.

What COX inhibitor, then, should be used to prevent the development of painful neuropathies? Selective COX-2 inhibitors have gained enormous publicity over the past years as a novel class of antiinflammatory and analgesic drugs with a largely reduced risk of upper gastrointestinal bleeding, which often limits the long-term use of classic (nonselective) cyclooxygenase inhibitors. 6The work by Hefferan et al.  2points to selective COX-1 inhibitors, which, by the way, may also exhibit reduced gastrointestinal toxicity. 7This question, however, is far from being settled. In the spinal cord dorsal horn, and in a number of other organs, including the kidney, COX-2 is already expressed at low levels under physiologic conditions but becomes dramatically increased after peripheral tissue inflammation. It is not clear why only COX-1 should contribute to painful neuropathy. In fact, if one looks carefully at the results by Hefferan et al. , 2it is clear that the selective COX-1 inhibitor SC-560 was less effective than the nonselective S-ibuprofen. Because no dose–response relationship has been performed, this interpretation must be made with caution. If it turns out to be true, one would therefore expect that a significant inhibition would also be likely after treatment with a selective COX-2 inhibitor. So, in a prospective trial, we would suggest comparing all three classes of COX inhibitors.

Two other unresolved questions are related to the pathophysiology of neuropathic pain. How do prostaglandins promote the development of painful neuropathies, and why are they only effective early in the course of the disease? Two recent publications have shed light on the molecular mechanisms of prostaglandin E2in the spinal cord. Baba et al.  8showed that prostaglandin E2directly depolarizes wide dynamic range neurons in the deep dorsal horn, and Ahmadi et al.  9demonstrated that prostaglandin E2reduces the inhibitory tone of the neurotransmitter glycine onto neurons in the superficial layers of the dorsal horn, thereby causing a disinhibition of spinal nociceptive transmission. Both mechanisms can explain why prostaglandin E2facilitates pain sensation. In addition, they may both contribute to plastic changes in neurotransmission between dorsal horn neurons, which may become prostaglandin-independent and largely irreversible during the disease course. In any case, the novel results point to a new possibility to prevent neuropathic pain, which, if already established, is largely refractory to current treatment options.

Zhu X, Eisenach JM: Cyclooxygenase-1 in the spinal cord is altered after peripheral nerve injury. A nesthesiology 2003; 99: 1175–9
Hefferan MP, O'Rielly DD, Loomis CW: Inhibition of spinal prostaglandin synthesis early after L5/L6 nerve ligation prevents the development of prostaglandin-dependent and prostaglandin-independent allodynia in the rat. A nesthesiology 2003; 99: 1180–8
Seltzer Z, Dubner R, Shir Y: A novel behavioral model of neuropathic pain disorders produced in rats by partial sciatic nerve injury. Pain 1990; 43: 205–18
Kim SH, Chung JM: An experimental model for the peripheral neuropathy produced by segmental spinal nerve ligation in the rat. Pain 1992; 50: 355–63
Scadding JW: Peripheral neuropathies, Textbook of Pain, 4th edition. Edited by Wall PD, Melzack R. Edinburgh, Churchill Livingstone, 1999, pp 815–34
Brune K, Zeilhofer HU: Antipyretic (non-narcotic) analgesics, Textbook of Pain, 4th edition. Edited by Wall PD, Melzack R. Edinburgh, Churchill Livingstone, 1999, pp 1139–53
Wallace JL, McKnight W, Reuter BK, Vergnolle N: NSAID-induced gastric damage in rats: Requirement for inhibition of both cyclooxygenase 1 and 2. Gastroenterology 2000; 119: 706–14
Baba H, Kohno T, Moore KA, Woolf CJ: Direct activation of rat spinal dorsal horn neurons by prostaglandin E2. J Neurosci 2001; 21: 1750–6
Ahmadi S, Lippross S, Neuhuber WL, Zeilhofer HU: PGE2selectively blocks inhibitory glycinergic neurotransmission onto rat superficial dorsal horn neurons. Nat Neurosci 2002; 5: 34–40