Review: Adverse Effects of Opioid Analgesics in Osteoarthritis Treatment: A Global Meta-Analysis, 2000–2022

Comprehensive computerized searches were conducted across multiple databases including PubMed, Embase, the Cochrane Library, Medline, Web of Science, and National International Trial Registers, complemented by key Chinese platforms such as China National Knowledge Infrastructure (CNKI), Wanfang Data, VIP Chinese Journal Service Platform (VIP CJSP), and Chinese Biomedical Literature Database (CBM). The study period extended from the inception of each database up to May 2024. There were no restrictions on the language of the literature. The search terms used were OA, knee arthritis, hip arthritis, opioids, RCT, and placebo. Details of the search strategies for each database are provided in Supplementary Table S1.

Inclusion criteria: 1) Studies were RCTs with no restrictions on language or publication status. 2) Participants were diagnosed with OA, with included subtypes such as hip and knee OA. 3) Interventions included opioids as a treatment for OA in the experimental group, with detailed reports on specific regimens (e.g., drug name, dose, treatment duration), and placebo administered to the control group. 4) Outcome measures focused on the type and frequency of opioid-related adverse drug reactions.

Exclusion criteria: 1) Non-clinical studies, case reports, reviews, comments, conference abstracts, and studies not addressing OA treatment; 2) Duplicate publications; 3) Studies with incomplete data reporting, which precluded a reasonable assessment of findings; 4) Studies classified as high risk in RCTs according to the Cochrane Risk of Bias tool, failing to meet a predefined quality threshold.

Two independent reviewers conducted a thorough literature screening, adhering stringently to predetermined inclusion and exclusion criteria. This process was followed by the creation of a standardized data extraction form to gather relevant information. Discrepancies were addressed by an expert committee, which reached a consensus depicted in Supplementary Figure S1. Data extraction from the selected studies included: 1) bibliographic information such as titles and authors; 2) study characteristics outlining research designs, sample populations, and intervention strategies; and 3) primary observational outcomes, focusing on the types and frequencies of opioid-related adverse events, as detailed in Supplementary Table S2.

The integrity and quality of the studies were rigorously assessed using the Cochrane Risk of Bias Tool. Analytical procedures were conducted using RevMan (version 5.4; Cochrane Collaboration, London, UK). Detailed experimental approaches are thoroughly described in the Supplementary Materials.

From an initial selection of 2,561 published studies in this field, 24 met the established inclusion criteria and involved a total of 9,586 OA patients. Of these participants, 4,782 were allocated to the opioid treatment group and 4,804 to the placebo control group. Analysis of gastrointestinal side effects revealed a higher incidence of nausea in the opioid group compared to placebo [relative risk (RR)=3.17, 95% confidence interval (CI): 2.83, 3.56, P<0.001] (Figure 1A). Similarly, constipation (RR=3.57, 95% CI: 3.15, 4.06, P<0.001) (Figure 1B) and vomiting (RR=3.65, 95% CI: 2.96, 4.49, P<0.001) (Supplementary Figure S2) were more prevalent in the opioid group. Dry mouth also occurred more frequently in this group (RR=4.14, 95% CI: 3.12, 5.50, P<0.001) (Supplementary Figure S3). However, the incidence of upper abdominal pain (RR=0.85, 95% CI: 0.50, 1.46, P=0.56) (Supplementary Figure S4) and diarrhea (RR=1.11, 95% CI: 0.90, 1.37, P=0.33) (Supplementary Figure S5) showed no significant differences between the two groups.

Figure 1. 

Occurrence of gastrointestinal-related adverse effects. (A) Meta-analysis of the incidence of nausea in the experimental (opioid treatment) and control (placebo) groups. (B) Analysis of constipation in both groups.

Abbreviation: CI=confidence interval.

We conducted an analysis of adverse reactions related to the nervous system, and general disorders at the administration site, as well as those involving skin, musculoskeletal, and connective tissues. Our results indicated a significantly higher incidence of dizziness (RR=3.06, 95% CI: 2.66, 3.52, P<0.001) (Figure 2A) and somnolence (RR=3.61, 95% CI: 3.01, 4.33, P<0.001) (Supplementary Figure S6) in the treatment group compared to the control group. Conversely, the difference in the incidence of headache was not statistically significant between the groups (RR=1.10, 95% CI: 0.97, 1.23, P=0.13) (Supplementary Figure S7). Fatigue occurrence was also higher in the treatment group (RR=2.52, 95% CI: 1.98, 3.22, P<0.001) (Figure 2B), along with incidences of hyperhidrosis (RR=4.85, 95% CI: 3.30, 7.13, P<0.001) (Figure 2C) and pruritus (RR=4.88, 95% CI: 3.70, 6.42, P<0.001) (Supplementary Figure S8). In contrast, the incidence of back pain was lower in the treatment group (RR=0.29, 95% CI: 0.15, 0.56, P=0.0002) (Figure 2D).

Figure 2. 

Occurrence of nervous system disorders, general disorders/administration site conditions, and adverse reactions related to skin, musculoskeletal, and connective tissue in the two study groups: (A) dizziness, (B) fatigue, (C) hyperhidrosis, (D) back pain.

Abbreviation: CI=confidence interval.

Our analysis revealed that the rate of treatment discontinuation due to intolerable drug-related adverse effects was significantly higher in the experimental group compared to the control group, with an RR of 6.00 (95% CI: 4.53, 7.95, P<0.001) (Figure 3A). Additionally, the incidence of total adverse events was also higher in the experimental group (RR=1.22, 95% CI: 1.14, 1.31, P<0.001) (Figure 3B). Specifically, the risk of experiencing severe adverse events was approximately 3.12 times higher in the experimental group than in the control group (95% CI: 1.65, 5.90, P<0.001) (Figure 3C). Severe adverse events identified in this study included respiratory depression, severe allergic reactions, intestinal obstruction from severe constipation, sedation-induced coma, cardiovascular incidents, and serious outcomes related to addiction. The criteria for classifying these events are detailed in the Supplementary Material.

Figure 3. 

Comparative analysis of adverse reaction incidences in the experimental and control groups. (A) Therapy discontinued. (B) Total adverse events. (C) Serious adverse events.

Abbreviation: CI=confidence interval.

This study, based on 24 RCTs, provides a detailed evaluation of the adverse effects associated with opioid analgesics for OA pain management. The results demonstrate that opioids effectively alleviate pain but also cause a range of adverse reactions. These include 1) Activation of central and gastrointestinal µ-opioid receptors, leading to delayed gastric emptying and vagal nerve activation, which may cause nausea and vomiting; 2) Reduced intestinal motility and decreased secretory activities, resulting in constipation; 3) Diminished central nervous system activity, which can cause sedation and drowsiness, potentially affecting cognitive functions, especially at higher doses or with long-term use (5). Additionally, a subgroup analysis involving three studies found a significantly lower incidence of back pain in the experimental group compared to the control group (P=0.0002). Although these results suggest that opioid treatment could reduce back pain in OA patients, the limited number of studies introduces the possibility of random error, limiting the generalizability of these findings to a larger OA patient population. The safety profile of opioid therapy for OA identified in this study is consistent with other research on opioid treatment for OA-related pain (6–8). The review of the included literature identified common opioids used for OA pain management in clinical practice, including Tapentadol ER, Oxycodone CR, Tramadol ER, and Hydrocodone ER. Standard dosages are 100 to 250 mg twice daily, 10–40 mg every 12 hours, 100–400 mg daily, and 15–90 mg every 12 hours, respectively. Some studies show that the incidence and types of medication-related adverse reactions vary with dosage (4–5,7).

While the primary focus of this study is on the adverse effects of opioids in managing OA, the significant risk of addiction associated with these drugs warrants attention. Prior research has revealed that the addiction rates for patients under long-term opioid therapy are considerably higher compared to other analgesics (9). Addiction is often driven by intricate neurobiochemical alterations, such as dysregulation in the dopaminergic system and disturbances in reward mechanisms (10). Specifically, the dosage and duration of opioid administration are directly linked to its addictive potential, with higher dosages and extended use markedly amplifying the risk of dependency and abuse (10). Additionally, certain individuals are more vulnerable to addiction due to genetic factors or a history of substance misuse. Based on the findings of our study, we strongly recommend that clinicians prescribe opioids according to the principle of using “the lowest effective dose for the shortest effective duration.” Moreover, we support the formulation and execution of a comprehensive multimodal treatment plan that includes physical therapy, psychological support, lifestyle changes, and suitable pharmacological treatments. This approach aims to comprehensively address pain and functional impairments, reduce medication-related side effects, and enhance overall disease management and quality of life.

The findings of this study have significant implications for policy-making and public health: 1) Drug regulatory agencies are urged to bolster the monitoring of opioids to curb abuse and illicit distribution. Additionally, the establishment of an exhaustive drug utilization registry system is essential for tracking and assessing opioid use and related adverse events. 2) Public health education efforts need to be escalated to enhance understanding of the benefits and risks associated with opioids, especially in community settings and primary healthcare facilities, to promote their judicious use. 3) It is advisable that local or regional pain management guidelines be revised and updated by the latest research findings to guarantee the safety and effectiveness of pain management practices.

The study is subject to some limitations: 1) The incompleteness of the dataset presents a significant challenge, as some findings are either unpublished or inaccessible, potentially leading to systematic biases. Future research should broaden the scope of database searches to ensure a more comprehensive and equitable data integration. 2) There is variability in adverse event reporting across studies, which often lack consistent detail in descriptions and classifications, impairing the accurate interpretation and comparison of safety data. Subsequent studies should conform to the guidelines established by international drug monitoring organizations or the World Health Organization to standardize the reporting of adverse events. 3) The duration of observation in some studies is relatively short (2–4 weeks), which restricts the evaluation of long-term adverse reactions, including addiction or other sustained effects. Future research should prioritize long-term follow-ups to assess drug safety fully.

Opioid analgesics provide transient relief from pain and enhance functionality in patients with OA. However, their severe adverse effects, such as pruritus, hyperhidrosis, dry mouth, vomiting, somnolence, constipation, nausea, dizziness, and fatigue, must not be underestimated. Moreover, prolonged usage carries risks of drug dependence and potential addiction. Therefore, it is imperative that future research focuses on assessing the long-term efficacy and safety of opioid analgesics while also developing strategies to minimize these negative outcomes. Employing more stringent research methodologies will improve our global understanding of the safety profile of these medications.

No conflicts of interest.