The Safety and Efficacy of Pre-and Post-Medication for Postoperative Endo-dontic Pain: A Systematic Review and Network Meta-analysis

: Background: Postoperative Endodontic Pain is a major concern for dentists and their patients, with pain having been reported to occur in 25%–40% of patients treated. Therefore, the aim of this systematic review and Network Meta-analysis (NMA) was to identify the safety and efficacy of pre-and post-medication for reducing postoperative endodontic pain. Methods: A literature search was performed in the SCOPUS, MEDLINE, and ScienceDirect, and Cochrane Central databases until December 2019 with no language restriction. Randomized controlled trials evaluating the efficacy of pre-or post-medications compared with other agents, placebo, or no treatment in adult patients who underwent endodontic surgery for postoperative pain were included. The mean difference of postoperative pain was measured using the Standardized Mean Difference (SMD) with its 95% confidence interval (95% CI).


INTRODUCTION
Postoperative pain during root canal therapy is a major undesirable complication for dentists and their patients.Anxiety and fear of pain during root canal treatment are the main reasons that prevent patients from attending dental offices [1].It was estimated that the prevalence of post-endodontic pain ranges from 3% to 58% [2 -4].This condition is linked with the exacerbation of inflammatory response and the activation of inflammatory mediators such as prostaglandins, which cause the periapical activation of sensitive nociceptors [5].Preoperative and procedural factors such as intracanal medicaments, mechanical instrumentation, microbial effects, and chemical irritants may cause periradicular tissue injury, which in turn causes post-endodontic pain [5,6].Endodontic treatment consists of restoring the form and function of teeth and controlling symptoms that address the primary concern of the patient as well as long-term possible complications, such as chronic pain [7].Therefore, it is highly important to manage discomfort during and after root canal treatment.
In this regard, many drugs have been used to relieve postendodontic pain, such as Non-steroidal Anti-inflammatory Drugs (NSAIDs), corticosteroids, opioids, cyclooxygenase-2 enzymes (COX-2) inhibitors, and combinations of drugs [8].Today, the most common types of pharmacological agents prescribed for pain relief in dentistry are NSAIDs and paracetamol (acetaminophen) [9].NSAIDs decrease inflammation, inhibit cyclooxygenase enzymes, and prevent new prostaglandin molecules, but have no effect on circulating molecules [10].Moreover, corticosteroids have demonstrated significant efficacy in dentistry pain management [11].Many randomized control trials were conducted to evaluate the efficacy of various oral pre-and post-medications such as prednisolone [12], ibuprofen [13], lornoxicam [14], indomethacin [15], gabapentin [14], and celecoxib [16].They reported that premedication is effective for postoperative pain after nonsurgical root canal treatment.However, the best painreducing agent is yet to be identified, as these drugs were not to be ranked regarding their efficacy.A recent network metaanalysis was conducted by Nagendrababu and his colleagues, who aimed to identify the most effective oral premedication in reducing pain in adults after nonsurgical root canal therapy [17].Nevertheless, their study failed to include all available evidence, which eventually affected their conclusion.In this systematic review and network meta-analysis, we aimed to summarize current evidence on the efficacy of pre-and postmedication for the treatment of postoperative endodontic pain and rank the available drugs according to their efficacy.

METHODS
This systematic review and network meta-analysis was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension statement for Network Meta-analyses of Health Care Interventions [18].

Risk of Bias Assessment
The revised Cochrane Collaboration's risk of Bias Assessment Tool (ROB) was used to assess the risk of bias among the included studies [19].Studies were evaluated for bias and categorized as having low, unknown, or a high risk of bias.The overall quality of the study was based on the 5 domains evaluated for bias: randomization, deviation from intended interventions, missing outcome data, outcome measurement, and selection of results.The overall score was low bias when all five domains were scored as low bias.The presence of at least two concerns in one of the domains rendered the study as having some concerns in bias.A study was evaluated as having high bias when at least one domain was scored to have high bias.

Data Synthesis and Statistical Analysis
The Standardized Mean Differences (SMD) in postoperative pain scores were calculated as the summary measures in MA.We chose SMD because changes in pain intensity scores were reported by different scales in trials, and the SMD can compare pain intensity scores in a uniform manner.In the case where variance data were not reported as standard deviation, it was estimated with algebraic recalculations or various approximation methods.Means and standard deviations were calculated from the reported medians, ranges, or Confidence Intervals (CIs) when not available.The presence of heterogeneity among the selected studies warranted the use of a random-effects model for the calculation of weighted Mean Differences (MDs) and 95% CIs in MA.The heterogeneity between trials was evaluated using I 2 statistics.Random effects NMA using a consistency model was applied to synthesize the available evidence by combining direct and indirect evidence from different studies.
The global inconsistency test using a fitting design-bytreatment model was used to identify the disagreement between the direct and indirect estimates as a measure of inconsistency.Frequentist method to rank treatments in network "netrank" function was used to rank the various interventions (the higher the P-score, the better the intervention).Moreover, the split direct and indirect evidence in network meta-analysis "netsplit" function was used.Publication bias was assessed using a comparison-adjusted funnel plot.All analyses were performed with R version 1.2.5019 (© 2009-2019 RStudio, Inc.) using the "netmeta" and "meta" packages for NMA [20].

Search Strategy Results
Our search retrieved 1512 unique citations.Following title and abstract screening, 107 full-text articles were retrieved and screened for eligibility.Of them, 45 articles were excluded, and 62 RCTs articles (n= 5412 patients) were included in the systematic review, and 50 articles were included in the final analysis.The flow diagram of study selection for our systematic review and meta-analysis is shown in PRISMA diagram (Fig. 1).A summary of included studies and baseline characteristics of the populations is shown in Table 1.

Characteristics and Quality of the Included Studies
A total of 5412 patients, including males and females between the ages of 15 and 80 years, from the included studies, formed the sample size for the NMA.The origin countries of included studies were Iran (n=21), USA (n=15), India (n=9), Egypt (n=3), Turkey (n=3), Brazil (n=2), Israel (n=1), Portugal (n=1), Sudan (n=1), and Australia (n=1), and four studies were found to be non-reported.Negm study consists of two trials; therefore, each one is considered as a separate study.The quality of the 62 included studies is described in Table 2. Thirty-seven studies had a low risk of bias, 10 studies had a high risk of bias, and 15 studies had some concerns.
Twelve Hours after Procedure: All medication showed a signіficant reduction when compared to placebo; Corticosteroids (SMD= -1.39), COX-2 inhibitors (SMD= -1.20), NSAIDs (SMD= -1.10), and Opioids (SMD= -0.84).Pooled analysis was heterogeneous (Q=507.44;I 2 =87.8%;P<0.0001) due to the significant variation among the analyzed categories (Fig. 2d).Publication bias analysis showed that there was a detected bias according to the Egger test (p=0.0001).Split analysis demonstrated that there was no significant difference between NSAIDs vs. corticosteroids, Opioids, and COX-2 inhibitors (Appendix Fig. 4).Network ranking graph showed the rank of categories at 12 hours after the procedure (Fig. 3d).Twenty-four Hours after Procedure: Among all medications, COX-2 inhibitors were ranked as the best treatment for the reduction of postoperative pain when compared to placebo [SMD=-1.27,95% CI (-2.10: -0.43), P-score=0.88].Corticosteroids and NSAIDs also showed a significant reduction in pain score (SMD= -1.13 and SMD= -0.65, respectively).Pooled analysis was heterogeneous (Q=81.07;I 2 =82.7%;P<0.0001) due to the significant variation among the analyzed categories (Fig. 2e).Publication bias analysis showed a detected bias according to the Egger test (p=0.0008).Split analysis demonstrated that there was no significant difference between NSAIDs vs. Opioids and COX-2 inhibitors (Appendix Fig. 5).Network ranking graph showed the rank of categories at 24 hours after the procedure (Fig. 3e).

League table is presented in Appendix Table 5.
Forty-eight Hours after Procedure: Among all medications, only NSAIDs demonstrated a significant reduction in postoperative pain when compared to placebo [SMD=-0.50,95% CI (-0.88: -0.13), P-score=0.76].Pooled analysis was heterogeneous (Q=129.7;I 2 =83.8%;P<0.0001) due to the significant variation among the analyzed categories (Fig. 2f).Publication bias analysis showed that there was no detected bias according to the Egger test (p=0.16).Split analysis demonstrated that there was no significant difference among NSAIDs, Corticosteroids or COX-2 inhibitors (Appendix Fig. 6).Network ranking graph displayed the rank of categories at 24 hours after the procedure (Fig. 3f).League table is presented in Appendix Table 6.

Postoperative Pain for Treatment Intervention Categorized by Chemical Name
Network diagrams of all the eligible comparisons for primary outcomes according to the chemical name are presented in Fig. (4a-f).
Immediately after procedure: Among all medications, Piroxicam was ranked as the best treatment for the reduction of postoperative pain [SMD= -1.20, 95% CI (-1.53: -0.86), P-score= 0.95].Moreover, Diclofenac sodium, Flubiprofen, Ketamin, Ketoprofen, and Ibuprofen showed a significant reduction in pain after endodontic treatment.Pooled analysis was found to be homogenous (Q=23.89;I 2 =20.5%;P<0.97) (Appendix Fig. 7).Publication bias analysis showed that there was no detected bias according to the Egger test (p=0.66).The split analysis is presented in Appendix Fig. (8).Network ranking graph showed the rank of drugs immediately after the procedure (Fig. 5a).
The split analysis is presented in Appendix Fig. (10).Network ranking graph showed the rank of drugs 6 hours after the procedure (Fig. 5b).
Eight hours after procedure: At this period, only four  11).Publication bias analysis showed that there was no detected bias according to Egger test (p<0.15).The split analysis is presented in Appendix Fig. (12).Network ranking graph showed the rank of drugs 8 hours after the procedure (Fig. 5c).
Twelve hours after procedure: Naproxen was ranked as the best treatment for the reduction of postoperative pain [SMD= -2.67, 95% CI (-3.90: -1.44), P-score= 0.92].Furthermore, Novafen, Indomethacin, Prednisolone, Gabapentin, Betamethasone, Dexamethasone, Rofecoxib, Piroxicam, and Ibuprofen showed a significant reduction in pain after 12 hours of endodontic treatment.Pooled analysis was found to be heterogeneous (Q=377.76;I 2 =86.8%;P<0.0001) due to the significant variation among the analyzed drugs (Appendix Fig. 13).Publication bias analysis showed that there was a detected bias according to the Egger test (p<0.0001).The split analysis is presented in Appendix Fig. (14).Network ranking graph showed the rank of drugs 12 hours after the procedure (Fig. 5d).
Twenty-four hours after procedure: Novafen was ranked as the best treatment for the reduction of postoperative pain [SMD= -2.13, 95% CI (-3.18: -1.08), P-score= 0.92].Furthermore, Naproxen, Indomethacin, Prednisolone, Gabapentin, Diclofenac sodium, Betamethasone, Dexamethasone, Rofecoxib, Kenacomb, Piroxicam, and Ibuprofen showed a significant reduction in pain after 24 hours of endodontic treatment.Pooled analysis was observed to be heterogeneous (Q=321; I 2 =84.4%;P<0.0001) due to the significant variation among the analyzed drugs (Appendix Fig. 15).Publication bias analysis showed that there was a detected bias according to the Egger test (p=0.003).The split analysis is presented in Appendix Fig. ( 16).Network ranking graph showed the rank of drugs 24 hours after the procedure (Fig. 5e).

Nausea
Our analysis showed that only five studies reported data regarding nausea [21 -25].Network graph included the following drugs: Indomethacin, ibuprofen, tramadol, betamethasone, flurbiprofen, and placebo (Appendix Fig. 19).Interestingly, among the tested drugs, no drug showed a significant increase in the risk/incidence of nausea, as shown in Appendix Fig. (20).Moreover, the ranking analysis demonstrated ibuprofen as the lowest drug associated with risk/incidence of nausea (Appendix Fig. 21).The split analysis is presented in Appendix Fig. (22).

Other Adverse Events
Salapoor et al.
[24] reported one case and Menhinick et al. [21]reported three cases of sweating due to using ibuprofen.Regarding dizziness, Shantiaee et al. [24] reported three cases with dexamethasone, and Sethi et al. [23]reported four cases with Tapentadol and Etodolac.In terms of vomiting and heartburn, three cases were recorded for each Tapentadol and Etodolac [23].

DISCUSSION
To the best of our knowledge, this is the largest and most updated systematic review and network meta-analysis that was conducted to evaluate the current evidence regarding the effect of pre-and postmedication for reducing the postendodontic pain.In this study, we included a total of 62 RCTs in the systematic review.Out of them, 50 studies were included in the network meta-analysis (NMA).NMA was conducted on the basis of pharmacological or chemical name groupings in order to identify the effect of classification of the medications given pre-or postendodontic care on postoperative pain during the following periods: immediately, 6, 8, 12, 24, 48 hours after the procedure.Opioids were ranked first in the pharmacologic group for reducing pain immediately after the procedure.Moreover, it showed a significant reduction at 12 hours after the procedure.Corticosteroids were ranked first as the best treatment for the reduction of postoperative pain at 6 and 12 hours with a significant reduction in postoperative pain scores [SMD= -1.18, 95% CI (-1.51: -0.85)] and [SMD= -1.39, 95% CI (-1.77: -1.02)], respectively.COX-2 were ranked as the best treatment for the reduction of postoperative pain at 8 and 24 hours with a significant reduction in postoperative pain scores [SMD= -2.86, 95% CI (-6.05: -1.66)] and [SMD= -1.27, 95% CI (-2.10: -0.43)], respectively.NSAIDs significantly reduced the postoperative pain scores in all durations.Based on the chemical name, piroxicam was superior immediately after the procedure, whereas indomethacin followed by novafen, naproxen, and prednisolone was found to be effective at 6 hours.At 12 and 24 hours, naproxen and Novafen followed by indomethacin were ranked first.However, at 48 hours, only indomethacin and betamethasone were effective.The safety profile of test drugs was acceptable except for some events of nausea, vomiting, and headache.
Clinically, it has been reported that patients with periapical diagnosis of an Acute Apical Periodontitis (APP) or Phoenix Abscess are more likely to require additional medication to relieve post-endodontic pain compared to a periapical diagnosis of a Normal Periapex, a Chronic Apical Periodontitis (CAP), or a Chronic Apical Abscess (CAA) [26,27].Therefore, it seems rational to minimize occlusion after root canal therapy on the tooth, which is harmful to percussion.Occlusal reduction in patients with teeth that initially show pulp vitality, percussion sensitivity, preoperative pain and/or absence of periradicular radiolucency has been recommended to prevent postoperative pain [28].On the other hand, CAA or CAP consists of a radiolucency at the root apex, a draining fistula (sinus tract), and usually no pain in percussion.Nagendrababu et al. [17] conducted NMA for the same purpose; however, they only included 16 RCTs and reported results for only three durations.In terms of adverse events, they reported a descriptive result and did not conduct a pooled analysis.In conclusion, they stated that the use of piroxicam or prednisolone would be the premedication of choice.We agree that these drugs are promising and show a significant effect; however, we believe that indomethacin, Novafen, naproxen, betamethasone have a better effect and longer duration.
In the NMA of Shirvani and colleagues, they aimed to investigate the efficacy of NSAIDs and paracetamol in reducing postendodontic pain.They did not include corticosteroids or opioids; therefore, they enrolled only 27 articles.They analyzed the data at four durations immediately, 6, 12, and 24 hours after the procedure.They performed a meta-regression which demonstrated that combination therapy did not reduce the pain significantly (OR= -0.88, 95% CI (-2.05, 0.28), p= 0.1).Moreover, they showed that the systemic administration was more efficient than oral administration (OR= -1.17, 95% CI (-1.93, -0.41), p= 0.004) and (OR= 4.24,95% CI (2.62,5.86),p<0.001), respectively.Finally, they recommended the use of multiple-dose regimens of NSAIDs during the postoperative period to achieve most efficacy (29).Smith et al. (30) found that the elimination of 6 hours of postendodontic pain with ibuprofen 600 mg and ibuprofen 600 mg + acetaminophen 1000 mg was more effective than placebo.They analyzed studies that evaluated the efficacy of pre-and postmedication for endodontic treatment on pain.They showed that ketoprofen 50 mg and naproxen 500 mg might be more effective than ibuprofen 600 mg at 6 hours postoperative.

Limitations
This study possessed some limitations: 1) Heterogeneity was observed in all analyses, which can be explained by the extensive variation in types of drugs, dosage, mechanism of action, and mode of administration.Moreover, the different types of teeth of participants with varied demographics may influence the applicability of our findings.However, all studies were conducted in hospitals, universities or clinics where the numbers and experience of operators were diversified, which could further encourage our findings to be generalized.2) We could not conduct a subgroup analysis according to the regimen doses because of insufficient data.

CONCLUSION
In conclusion, the current evidence suggests that pre-and postmedication have the ability to reduce postoperative pain after nonsurgical root canal treatment.Corticosteroids and COX-2 inhibitors showed significant control of the pain up to 12 hours after administration.However, NSAIDs demonstrated a high efficacy from administration and until two days after treatment.Indomethacin, Novafen, prednisolone, and Naproxen were ranked as first in most analyzed durations.The use of narcotic agents before and post-nonsurgical root canal procedures for postoperative pain control and improving the quality of life needs further research.

CONSENT FOR PUBLICATION
Not applicable.

FUNDING
None.

Fig
Fig. (2).Forest plot of the effect of Treatment Intervention Categorized by Pharmacologic Group on Postoperative Pain: a) Immediately after Procedure, b) Six Hours after Procedure, c) Eight Hours after Procedure.d) Twelve Hours after Procedure, e) Twenty-four Hours after Procedure, f) Forty-eight Hours after Procedure.

Fig. ( 3
Fig. (3).Network ranking graph showed the rank of categories for the primary analysis Categorized by Pharmacologic Group: a) Immediately after procedure, b) Six Hours after Procedure, c) Eight Hours after Procedure.d) Twelve Hours after Procedure, e) Twenty-four Hours after Procedure, f) Forty-eight Hours after Procedure.

Fig
Fig. (4).Network diagrams of all the eligible comparisons for primary outcomes according to the chemical name: a) Immediately after procedure, b) Six Hours after Procedure, c) Eight Hours after Procedure.d) Twelve Hours after Procedure, e) Twenty-four Hours after Procedure, f) Forty-eight Hours after Procedure.

Fig. ( 5
Fig. (5).Network ranking graph showed the rank of categories for the primary analysis Categorized by the chemical name: a) Immediately after procedure, b) Six Hours after Procedure, c) Eight Hours after Procedure.d) Twelve Hours after Procedure, e) Twenty-four Hours after Procedure, f) Forty-eight Hours after Procedure.