Thoracotomy is often responsible for chronic pain, possibly of neuropathic origin. To confirm preclinical studies, the preventive effects of perioperative ketamine were tested in a randomized, double-blind, placebo-controlled clinical trial on persistent neuropathic pain after thoracotomy. Eighty-six patients scheduled for thoracotomy under standardised general anaesthesia were randomised to receive either ketamine (1 mg kg(-1) at the induction, 1 mg kg(-1) h(-1) during surgery, then 1 mg kg(-1) during 24 h; n=42) or normal saline (n=44). Postoperative analgesia included a single dose of intrapleural ropivacaine, intravenous paracetamol and nefopam, and patient-controlled intravenous morphine. Vital parameters and analgesia were recorded during the 48 first postoperative hours. Seventy-three patients were followed up. The patient's chest was examined 1-2 weeks, 6 weeks and 4 months after surgery. At the last two observations, spontaneous pain score over a one-week period (visual analogue scale), neuropathic pain score (NPSI), and intake of analgesics, were assessed. No drug affecting neuropathic pain (except opiates) was given during the follow-up. Two patients in each group were lost to follow-up after the 6 week visit. Ketamine improved immediate postoperative pain, but the groups were similar in terms of neuropathic pain and intake of analgesics, 6 weeks (NPSI score: ketamine: 1.25 [0-4.125]; placebo: 1 [0-4]) and 4 months after surgery. Thus, ketamine given in 24-h infusion failed to prevent chronic neuropathic pain after thoracotomy. Other perioperative preventive long-lasting treatments or techniques could be tested in this context.

This study evaluated prospectively the incidence of neuropathic pain after thoracotomy, described its clinical characteristics, and delineated landmarks for its diagnosis in daily practice. We evaluated clinically painful symptoms and sensory deficits in 54 patients after lateral/posterolateral thoracotomy for broncho-pulmonary carcinoma with standardized surgical and analgesic procedures. At 2months, 49 patients suffered from non malignant thoracic pain, and at 6months 38 patients (loss to follow-up for 7) reported persisting pain. In 35 patients, painful symptoms and sensory deficits could be evaluated using a standardized clinical bedside procedure. According to the grading system proposed by Treede et al. [41], neuropathic pain was considered probable in 21 patients, while use of the DN4 questionnaire concluded that neuropathic pain was probable in 17 patients. The two diagnostic procedures provided similar conclusions in 16 patients. Morphine consumption during the early post-operative period (mean 111.3±30.8mg/day) and pain intensity (VAS: mean 5.71±2.1) were significantly higher in patients suffering from neuropathic pain than in other patients with pain (mean 80±21.4mg/day; VAS: mean 3.9±2.4). The clinical picture in most patients with neuropathic pain included electric shocks and severe multimodal hypoesthesia in the sensory area of 5th/6th intercostal nerves. Thus, our results indicate a minimal incidence of chronic post-thoracotomy pain at 70% and that of neuropathic pain at 29%, this latter being clinically suggested by a combination of certain symptoms and reinforced by the DN4 questionnaire when sensory deficit at scar is present.

Background: The opioid use disorder is an international public health problem. Over the past 20 years it has been the subject of numerous publications concerning patients treated for chronic pain other than cancer-related. Patients with cancer-related pain are also at risk of opioid use disorder. The primary objective of this literature review was to determine the prevalence of opioid use disorder in patients with cancer-related chronic pain. Its secondary objective was to identify the characteristics of these opioid users. Methods: This is a literature review of studies published over the last twenty years, from 1 January 2000 to 31 December 2020 identified by searching the three main medical databases: Pubmed, Cochrane, and Embase. A meta-analysis took account of between and within-study variability with the use of random-effects models estimated by the DerSimonian and Laird method. Results: The prevalence of opioid use disorder was 8% (1–20%) and of the risk of use disorder was 23.5% (19.5–27.8%) with I2 values of 97.8% and 88.7%, respectively. Conclusions: Further studies are now needed on the prevalence of opioid use disorder in patients treated for cancer-related chronic pain. A screening scale adapted to this patient population is urgently needed.