Pain is one of the most prevalent and costly health conditions. The estimated cost of pain in the United States was more than that of heart disease and cancer treatments, reaching $560 billion to $635 billion per year (1). Acute abdominal pain is the leading symptom and principal reason for hospital admission in patients with acute pancreatitis (AP).

Acute pancreatitis (AP) is one of the most common gastrointestinal diseases with an increasing global incidence (2), for which there is no specific targeted therapy (3). According to the latest epidemiological investigation, aggregate annual health care cost of AP rose to $2.6 billion in the United States in 2014 (4) and £200 million in the United Kingdom (NHS). Abdominal pain not only serves as one of the diagnostic criteria (5), but also a prognostic factor (6), and is related to the length of hospital stay and other self-reported outcomes of patients with AP (7, 8).

Nearly all patients with AP experienced abdominal pain which warrants prompt analgesia, and this is one of the main management priorities in the early management of AP (9). Some current practise guidelines have overlooked recommendations for pain management (10–14), while others provide a clear recommendation for pain assessment (15, 16), and best pharmacological options (15) (Table 1). No guidelines provide sufficient details regarding the type, dose, route, and frequency of analgesia administration.

There are many pharmacological options, with opioids being the most frequently prescribed analgesics for pain relief of patients with AP. Since abdominal pain in AP is secondary to pancreatic parenchymal inflammation (17, 18), non-steroidal anti-inflammatory drugs (NSAIDs) that target the enzyme cyclooxygenase (COX) are often used (19, 20). Much less frequently, local anaesthetics (i.e., procaine and bupivacaine) and paracetamol (19–21) are used.

A previous meta-analysis of only 5 studies (22) (227 participants) showed there was no significant difference between opioids vs. non-opioids (3 studies, 162 participants) regarding the need for rescue analgesia, pain intensity, clinical outcomes, and adverse events for pain relief of patients with AP. It is noted that opioid requirement during hospitalisation is strongly associated with more complications of this disease (23, 24) and there is the need for more evidence regarding the safety and efficacy of individual opioids in AP management. Further, a recent systematic review (25) demonstrated that NSAIDs reduced AP severity and mortality rate in animal models and patients. However, this review included studies that were of poor quality, the efficacy of NSAIDs for pain relief in this setting still needs to be addressed. Recently, thoracic epidural anaesthesia has been shown to significantly improve pancreatic microcirculation and splanchnic perfusion and reduce indices for multiple organ failure and mortality (26). There is the need to repeat a systematic review on this topic and it includes recent clinical studies of pain relief protocols in patients with AP to guide clinical management and trials.

The aim of this systematic review and meta-analysis was to include recent randomised clinical trials (RCTs) to evaluate the safety and efficacy of analgesics in the management of pain in patients with AP.

The study was conducted following protocols of the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) Statement (27).

This systematic review and meta-analysis included 12 RCTs investigating analgesics for pain relief in AP and comprised a total of 699 patients, of whom 83% were mild cases. It was found that the use of analgesics was associated with reduced need for rescue analgesia and with pain score compared with controls. Unlike the previous meta-analysis (22) showing opioids were equivalent to non-opioids in decreasing the need for rescue analgesia, this meta-analysis found that opioids were superior to non-opioids in regards to the need for rescue analgesia but without significant differences in pain score, other clinical outcomes, and adverse events. It was found that NSAIDs and opioids were equally effective in decreasing the need for rescue analgesia in patients with AP, which was not the case for the systemic administration of local anaesthetics. The lack of heterogeneity in these two subgroups analyses makes the conclusion more convincing.

Of the 4 RCTs that investigated the effects of analgesics in comparison with controls, 2 reported that NSAIDs reduced the need for rescue analgesia. Ebbehoj et al. (34) showed that indomethacin was effective for pain relief. In a most recent study, Huang et al. (35) demonstrated that parecoxib plus celecoxib (selective COX-2 inhibitors) was associated with significantly less meperidine injections, reduced pain score, systemic and local complications, length of hospital stay, and cost without evidence of adverse events. Indeed, NSAIDs (i.e., celecoxib) are strongly recommended by the EARS Society as a component of a multimodal analgesic strategy and have been shown to reduce PCIA morphine consumption after major surgery in patients without contraindications (45, 46). Overall, our findings support the use of NSAIDs (in the absence of acute kidney injury) for pain relief in AP which is not a feature in previous AP guidelines.

Compared with the Cochrane database review conducted by Basurto et al. (22) in 2013 which included 5 RCTs with 227 participants, the current systematic review included 12 RCTs with 699 participants. For opioids vs. non-opioids, while Basurto et al. (22) did not show significance (3 studies with 162 patients, OR 0.41, 95% CI 0.14 to 1.19, p = 0.10), our results of an additional 3 RCTs favoured opioids over non-opioids (6 studies with 338 patients; OR 0.25, 95% CI 0.07 to 0.86, p = 0.03) in reducing the need for rescue analgesia. Subgroup analysis of 2 studies comparing opioids with procaine did not change the result of the primary outcome (OR 0.02, 95% CI 0.00 to 0.08, p < 0.00001), indicating opioids are a more effective analgesic than the systemic administration of procaine. The rationale for using procaine is because of its local analgesic effect and, possibly, its inhibitory effect on phospholipase A2 catalytic activity, which is an important enzyme involved in the early pathogenesis of AP (47). Therefore, systemic administration of local anaesthetics, mainly procaine, was already recommended as front-line analgesic in AP by consensus conferences in several European countries two decades ago (48, 49). Our results suggested that systemic use of local anaesthetics is not supported and, thus, its role in pain management in AP needs to be defined.

Subgroup analysis of 4 studies between opioids and NSAIDs revealed there was no significant difference in the need for additional analgesic (OR 0.56, 95% CI 0.24 to 1.32, p = 0.18) without heterogeneity (I² = 0), Δ-VAS, other clinical outcomes, and adverse events. Recently, a retrospective study conducted by Kim et al. (50) showed that opioid was associated with an increased risk of developing AP compared to NSAIDs (OR 2.64, 95% CI 1.54 to 4.52). It is reported that opioids can potentially cause AP in patients with a history of cholecystectomy due to its adverse effect of sphincter of Oddi constriction (50, 51). There are other effects of opioids including ileus, dysbiosis, opioid hyperalgesia, and others (22, 52). However, opioids have been overused for pain management in AP especially in North America where 92.5% of patients with AP received opioids for impatient pain management and 64.3% have been given opioids at discharge (53). On the other hand, the over prescription or prescription of opioids without adequate supervision has led to an alarming rise of opioid overdose-associated deaths (54). From 2013 to 2019, the synthetic opioid-associated death rate increased more than 10 folds, from 1.0 to 11.4 per 100,000 age-adjusted population in the United States (55). Therefore, based on the available evidence, it appears that NSAIDs are preferred over opioids as first line analgesia in patients with AP.

Epidural analgesia is an essential component of most enhanced recovery after surgery (ERAS) pathways and is commonly used in major abdominal surgery, because it is associated with superior pain control (46). However, epidural analgesia is rarely used in patients with AP. Sasabuchi et al. (56) investigated 44,146 patients with AP in Japan between 2010 and 2013 and found that only 0.7% of patients received epidural analgesia for pain management. A large propensity score-matched retrospective observational study (57) of 1,003 patients conducted in 2018 demonstrated that the mortality of critically ill patients with AP who received epidural analgesia was significantly lower than those who did not. In our study, there was only one RCT focusing on PCEA vs. PCIA (37). It was clear that PCEA markedly improved pancreatic arterial perfusion over PCIA, and this finding is consistent with studies from experimental AP models (26) and with patients who are critically ill with AP (37). Meanwhile, a multi-centre RCT (58) is underway to elucidate the benefits of PCEA among critically ill patients with AP. PCEA is not currently recommended for patients with in mild and moderately severe AP because of potential adverse effects including catheter placement-related hypotension and epidural abscess, although at a relatively low incidence. More studies evaluating the safety and efficacy of epidural analgesia in patients with severe AP are warranted.

In our study, we did not include studies assessing the analgesic effect of acupuncture on patients with AP as we focused on the pharmacological intervention. A systematic review and meta-analysis (59) have demonstrated that acupuncture was associated with significantly reduced abdominal pain (mean difference −0.87, 95% CI −1.01 to −0.73, p < 0.00001), improved gastrointestinal function, accelerated time of resuming to diets, and shortened the length of hospital stay without noticeable adverse events. However, most of these studies are published in Chinese and more rigorously designed RCTs are needed to confirm these findings.

It is striking that there is not more level 1 evidence available to guide decision-making with optimised analgesic protocols in patients with AP. The evidence available from the 12 RCTs is limited by the quality and heterogeneity of included studies as shown by our sensitivity analysis. In order to minimise the impact of heterogeneity, the random-effects model was adopted in the quantitative analysis for all outcomes. The unblinding characteristic of some studies may influence the patient-reported outcomes including the scoring of pain. Moreover, the variation in drug administration route and dosages impacted the study outcomes and this is reflected in the low GRADE levels. For example, in opioids vs. non-opioids, Peiró et al. (43) utilised smaller doses of morphine subcutaneously while other studies administered greater doses of opioids by intravenous approach. This might also explain the differences between studies. Other aspects that impact the trial design and outcomes are differences in the timing of pain onset in relation to hospital admission or recruitment, admission pain intensity, and predicted severity. We chose the primary outcome to be “the need for rescue analgesia” as data were not available on opioid equivalent for the rescue analgesic in most studies. For future trials, there is also the need for standardised reporting of pain outcomes and inclusion the identification of objective biomarkers (electrophysiology, bioassay, omics, imaging, and behaviour) (54).

The important finding from this study is that NSAIDs are as effective as opioids and opioids are more effective than systemic local anaesthetics on reducing the need for rescue analgesia in patients with mild AP. Epidural anaesthesia shows promise but will need further trials in patients with predicted severe AP before it can be recommended or adopted.

The original contributions presented in the study are included in the article/Supplementary Materials, further inquiries can be directed to the corresponding authors.

WC, FL, and YW are joint first authors of this article, undertook the statistical analysis, and interpreted data. WH conceived and designed the study. WC, FL, YW, and CH undertook the systematic literature search, acquisition of data, quality cheque, and risk of bias assessment. MP contributed to the design of statistical analysis. WC, FL, YW, and WH drafted the manuscript. WH and QX obtained funding and supervised the study. VS and RS had important intellectual input. All authors have approved the final version of manuscript before submission.

This project is supported by the National Nature Science Foundation of China (No. 82104598, CH; No. 81973632, WH). NZ-China Strategic Research Alliance 2016 Award (China: 2016YFE0101800, QX and WH). WC is funded by China Scholarship Council (No. 201806240274) and West China Hospital-Liverpool Clinician Scientist Development Award.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.