Although well intended, surgical procedures can result in chronic postsurgical pain (CPSP). Moreover, the treatment of postsurgical pain with opioids may lead to persistent postoperative opioid use (PPOU), opioid-use disorder (OUD), and opioid-related harm to patients and households. It is therefore vital to mitigate CPSP and opioid use after surgery, and regional anesthesia has long been considered a logical tool to do so.
In this issue of Anesthesia & Analgesia, Pepper et al present a meta-analysis of randomized trials evaluating the effects of regional anesthesia on CPSP and PPOU. Studying these outcomes is complicated by multiple factors, several of which are presented below. It may be telling to first discuss some of the limitations of the analysis by Pepper et al. The authors initially set out to identify and to aggregate trials that evaluated the single primary outcome of PPOU, but they identified only 5 such studies. Among the 5 studies, the definition of PPOU varied widely, and the studies did not actually measure opioid use. Instead, all 5 studies surveyed patients about analgesic use, which may have included or solely represented nonopioid analgesics. Most of the included studies defined analgesic use as any amount being taken at the time of follow-up. The regional anesthesia interventions studied also varied considerably (3 epidural infusion, 1 femoral nerve block, and 1 paravertebral block), as did the time points measured (3, 6, or 12 months after surgery). Pepper et al found no difference in the incidence of PPOU (actually, analgesic use) at any one of the individual measured timepoints: 3, 6, or 12 months, and they report a significant effect that was only seen when data from all 3 timepoints were pooled together. This resulted in an effect size just barely crossing the threshold of statistical significance (P = .04). The included studies were assessed as low quality using the GRADE tool. Furthermore, 3 of 5 studies had a high risk of bias using the Cochrane Risk of Bias 2 tool.
Given the paucity of studies assessing PPOU in their initial search, Pepper et al broadened their primary outcomes to include CPSP. They identified a total of 26 trials that evaluated CPSP in some form, though the definition of CPSP varied widely across the trials. For this outcome, they analyzed each measured timepoint independently and reported a significant decrease in CPSP at 3 and 6 months when regional anesthesia was used, but not at 12 months. While their data for the outcome of PPOU were based on a total of only 348 patients, the pooled data on CPSP ranged from approximately 700 to 1800 patients in each arm. The strongest evidence was for a decrease in CPSP at 6 months after surgery, with a total of 3457 patients across 19 pooled studies, reaching a high level of statistical significance (P < .0001) and graded as moderate in quality.
Pepper et al’s findings contrast somewhat with those of Weinstein et al’s Cochrane review and meta-analysis of 39 randomized clinical trials, which showed a lower incidence of CPSP for up to 18 months when regional anesthesia was used after thoracotomy and for up to 12 months after cesarean delivery or breast surgery. Considering the potential etiologies for this discrepancy reveals many of the problems inherent in studying chronic pain outcomes. First, the study by Pepper et al may have been underpowered to find such a difference at 12 months, which is indicative of the need for more high-quality trials in transitional pain medicine. Another possibility is that the study populations, regional interventions, or surgical procedures may have been different between the 2 meta-analyses.
Given that surgical procedures vary widely, some are associated with more severe and longer pain trajectories than others. While regional anesthesia modalities are well-known to be opioid-sparing in the intraoperative and postoperative periods, studies of whether such opioid-sparing techniques translate to lower incidences of PPOU or CPSP have been thus far inconclusive. Existing studies of CPSP favor the use of thoracic epidural analgesia for preventing chronic post-thoracotomy pain, neuraxial anesthesia for chronic postcesarean section pain, and paravertebral blocks for chronic post-mastectomy pain,5,6 but they do not support the use of peripheral nerve blocks for preventing chronic pain after upper and lower extremity surgeries.
Similarly mixed results have been observed in studies on the use of regional anesthesia to prevent PPOU. This may be due to a lack of prolonged opioid-sparing effect with regional anesthesia, the influence of enhanced recovery protocols or the reflection of a general trend toward lower opioid prescribing practices in the last decade. Furthermore, a narrower scope of surgical procedures may reveal a specific effect that is no longer seen when pooled with data from other interventions. Some of the discrepancies may also be due to inconsistencies in the measurement of CPSP. For example, a study using a very narrow CPSP definition may find a small treatment effect, while a study using a looser definition of CPSP in the same population may find a large treatment effect. It follows that if we are to make meaningful inferences about the impact of regional anesthesia on the development of chronic pain after surgery, we either need an extremely large amount of patient data that can be pooled to eliminate the effect of variable CPSP definitions, or we need to study one standard set of CPSP outcomes across all studies.
The need to verify a direct relationship with the index surgery further complicates the study of CPSP and PPOU. In some instances, patients taking opioids at 12 months after surgery may be doing so for reasons other than the surgery that is being investigated. As Jivraj et al astutely noted in their editorial, opioid use at 12 months after surgery may be caused by a new painful medical problem (51%) or another more recent surgery (40%).
In addition to causality, pathophysiology should also be considered in defining CPSP. Neuropathic processes facilitate the transformation of acute postoperative pain to CPSP, and in some patient cases, neuropathic pain can be a major component of CPSP. Identification of CPSP with a predominantly neuropathic component can be achieved using a validated questionnaire such as the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS), the Neuropathic Pain Questionnaire (NPQ), or the neuropathic pain diagnostic questionnaire (DN4). Recognition and treatment of neuropathic pain is critical as it may affect the long-term pain trajectory and may warrant the use of certain nonopioid analgesics over opioids.
From the discussion so far, the study of CPSP clearly presents challenges as well as opportunities. Investigations of CPSP tend to use 2 main outcome domains: (1) the presence of pain itself, or (2) the surrogate measure of opioid use (PPOU). Tools for measuring CPSP and the timing of its measurement have been the subject of several publications. However, it is notable that the definition of PPOU, which may seem more objective and measurable than CPSP, could be even less standardized than that of CPSP. A 2020 systematic literature search found 29 different definitions of PPOU and an incidence that varied 100-fold depending on the definition used. Definitions varied in (1) the amount of opioid: anywhere from one prescription to continuous use, (2) whether a prescription was filled or actually consumed, and (3) the timeframe of use: anywhere from 30 to 360 days after surgery. A movement toward standardization and consistency in outcome reporting could considerably improve our ability to draw inferences from longitudinal studies and from pooled data.
The development of multidimensional outcome measures is another future refinement related to CPSP. The 2005 Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) recommendations define outcome domains and tools for the study of chronic pain, which include pain, physical functioning, emotional functioning, side effects, and patient satisfaction. Subsequently, multidimensional outcome measures of perioperative pain have been developed. These include the International Pain Outcomes Questionnaire (IPO) and the Quality of Recovery scoring system for postoperative pain. Regarding studies on CPSP, the inclusion of measures of function is one recent advance within a composite patient-reported outcome pain score.
Hofer and colleagues used a pain composite score consisting of worst pain, least pain, and percent of time in severe pain; and a pain interference total score that noted pain interference with activities in bed, deep breath, coughing, sleep, anxiety, and helplessness. It is notable that the scoring system considers functional and emotional interferences. Hofer et al then categorized patients into 3 groups with well-defined cutoffs: no CPSPF, CPSPF, and mixed (intermediate). The psychometric properties of this measure and future tools need to be tested and validated, and a measure of patient satisfaction ideally added to fully represent all domains advocated by the IMMPACT group. With these modifications, tools like CPSPF could represent a viable candidate for a shared and standardized outcome measure to study CPSP across all populations.
As anesthesiologists dedicated to enhancing patient outcomes, we must use rigorous and high-quality research methodologies when performing individual trials and when pooling their results into meta-analyses. Such efforts begin with standardizing the definition of outcomes (eg, CPSP and PPOU), and follow with generating high-quality evidence primarily aimed at such outcomes. While the findings by Pepper et al4 presented in this issue of Anesthesia & Analgesia add some evidence that regional anesthesia may play a role in reducing the incidence of chronic pain after surgery, the evidence remains limited and difficult to generalize to all regional anesthetic techniques. Readers should thus be cautioned against interpreting these findings as conclusive. The relatively small numbers of patients for some outcomes, heterogenous interventions, variable outcome measures, and mixed quality of included studies warrant this caution. More conclusive evidence requires confirmation of the relationship of CPSP or PPOU to the index surgery, identification of the nature of the CPSP, collective standardization of outcomes, large multi-center randomized trials using standardized interventions and standardized outcomes, and/or novel methods to glean more granular information from large data-driven nonrandomized studies. The development and use of standardized multidimensional outcome measures in future studies has great potential to enhance our understanding of the effect of regional anesthesia interventions on CPSP.
Leave a Reply
You must be logged in to post a comment.