Persistent postoperative pain after ambulatory paediatric surgery

Editor, The literature on persistent postoperative pain (PPP) in paediatrics is very limited, with only a few prospective studies.1,2 Only one study has explored PPP in ambulatory children but it failed to specify the risk factors.3 Owing to the frequent neuropathic characteristics of PPP, the diagnostic method used in adults is the DN4 questionnaire,4 but this is not suitable for a paediatric population. The main objective of this prospective observational study was to evaluate the incidence of PPP 3 months after outpatient surgery in children aged 6 months to 15 years, using an electronic DN4 questionnaire modified for children. The secondary objective was to identify risk factors associated with PPP. The single-centre study was conducted in a Paediatric University Hospital, after approval from our Review Institutional Board. The survey included outpatients scheduled for visceral, urological, orthopaedic or reconstructive plastic surgery, after parental informed written consent. The choice of anaesthetic and postoperative pain treatment was left to the discretion of the anaesthesiologist. Data on patient characteristics, anaesthesia consultation, intra-operative period, recovery room stay and postoperative analgesic prescriptions were recorded using our institutional anaesthesia software, Pegasus version 4.3.0.r1 (Télème; Angers Technopole, Toulouse, France). A questionnaire derived from the DN4 questionnaire,4 was created in Google forms and a link emailed to the parents 3 months after the surgery. The eight DN4 items detecting neuropathic signs were reformulated with a child's language. The additional two items that required clinical examination were replaced by a query as to the presence of any pain experienced at the surgical site, and on the disabling or not character of this pain. For children under 5 years of age, the questionnaire was completed by the parents. Pain was considered as persistent when at least one positive response to any of the 10 questions was obtained. Statistical analysis was performed using the Stat View software (Abacus Concepts Inc. Berkeley, California, USA). Data were medians [IQR] or number (%) of responders and were analysed using non parametric tests or contingency tables. The risk factors associated with PPP in the responder population were identified with a univariate analysis followed by a logistic regression multivariate analysis. The analysis included 453 children, with 168 (37%) responding to the 3-month questionnaire. Patient characteristics and anaesthetic and surgical management of the responder and nonresponder populations were not different. Age of responders was 60 [36 to 108] months with 52.4% of them less than 6 years of age and 17.3% under 2 years of age. Pre-operative pain at the surgical site was present in 1.8% of cases. Regional anaesthesia was performed in 40.5% of children and sufentanil was administered in 78.6% of cases. Children also received paracetamol (100%), ketamine (17.3%), ketoprofen (60.7%), dexamethasone (11.9%) and ondansetron (11.9%). Anaesthesia and surgery duration were, respectively, 54 [36 to 72] and 21 [13 to 32] min. Intravenous morphine was given in the recovery room to 12.5% of the children, with a median dose of 88 [83 to 103] μkg−1. There were 49 (29.2%) questionnaires with at least one positive response: 4, burning; 9, painful cold; 5, electric shock; 5, paraesthesia; 20, tingling; 22, itching; 24, painful friction; 16, any pain, and 6, disabling pain. There were 11 (6.5%) questionnaires with four positive responses. Pain was reported without neuropathic signs in three (1.8%) questionnaires. In 32 cases there were two [1 to 3] neuropathic signs [range: 1 to 6] but without pain at the surgical site. Surgery was orthopaedic equipment removal, epiphysiodesis, ligamentoplasty, cryptorchidism, circumcision, inguinal or umbilical hernia, hypospadias, and plastic surgery. Multivariate analysis identified two independent risk factors for PPP: age at least 72 months at the time of surgery [odds ratio (OR), 2.6 95%CI 1.29 to 5.33) P = 0.008] and the need for morphine titration in the recovery room [OR, 3.5 (95% CI 1.34 to 9.28) P = 0.01] with an area under the curve = 0.64 (χ2 Hosmer and Lemeshow = 0.22, Pearson coefficient = 0.96). At the 35.3% threshold, sensitivity was 71.4%, specificity 51.7%, positive predictive value 38.0%, negative predictive value 81.3%. The positive and negative likelihood ratios were 1.53 and 0.55, respectively. Statistical analysis failed to identify the independent factors associated with four positive responses to the questionnaire except, again, for age. Our major finding is the high incidence of persistent neuropathic signs after paediatric outpatient surgery. Knowledge of risk factors is essential to identify the target for interventions to improve long-term postoperative recovery. The need for morphine titration in the recovery room probably reflects acute postoperative pain: such postoperative pain has been linked to PPP in children after thoracotomy.2 In adults, prevention of PPP is based on multimodal analgesia, including, if possible, a regional anaesthetic technique and ketamine. Our study, like others, failed to demonstrate a favourable effect of these two techniques in paediatrics. The second risk factor was age over 6 years. Greater neuronal plasticity and flexibility of musculoskeletal structures along with the absence of fear regarding surgery in younger children could explain this. Another possible explanation is the lack of sensitivity of neuropathic signs in the diagnosis of PPP in young children. Given this possible difficulty, we considered PPP to be positive from a single response, while in adults a DN4 questionnaire is positive with four responses.4 Further studies should consider other factors that could also contribute to PPP, for example psychosocial factors such as parental and child catastrophism and anxiety as well as poor compliance with home analgesic treatment. Our results justify paying particular attention to postoperative pain control to optimise the long-term rehabilitation of our young people after minor surgery. Acknowledgements relating to this article Assistance with the study: none. Financial support and sponsorship: none. Conflict of interest: none.