Efficacy of 4% articaine vs 2% lidocaine in mandibular and maxillary block and infiltration anaesthesia in patients with irreversible pulpitis: a systematic review and meta-analysis

Objective The goal of this systematic review and meta-analysis is to determine the performance of 4% Articaine vs. 2% Lidocaine for mandibular and maxillary block and infiltration anaesthesia in patients with irreversible pulpitis (IP). Methods PubMed/MEDLINE, Cochrane Central Register of Controlled Trials, Web of Science, Google Scholar, and Open Gray were used to conduct a thorough literature search. A manual search of the reference lists of the publications found was also carried out. Two reviewers critically evaluated the papers for inclusion and exclusion criteria, and data extraction was done on the selected publications. The Cochrane Collaboration Tool and the Minors checklist were used to assess the quality of the selected studies for randomised controlled trials (RCTs) and non-randomised studies, respectively. The RevMan software was used to perform a meta-analysis of the pooled data and subgroups according to the technique of anaesthetic solution delivery, as well as a sensitivity analysis (P < 0.05). Results A total of twenty-six papers were included in the qualitative synthesis, with twenty-two of them being included in the meta-analysis. There were fifteen studies with a low potential for bias, three with a moderate potential for bias, and seven with a high potential for bias. The combined results of the 19 trials in the tooth level unit revealed that 4% articaine had a success rate 1.37 times greater than 2% lidocaine for mandibular teeth (RR, 1.37; 95% CI [1.17–1.62]; P = 0.0002). For the maxillary buccal infiltration method, the combined results from the three trials revealed that 4% articaine resulted in a success rate 1.06 times greater than 2% lidocaine (RR, 1.06; 95% CI [0.95–1.2]; P = 0.3). Excluding subgroups with a single study in sensitivity analysis for mandibular teeth revealed a substantial improvement in the success rate of the articaine group in treating IP when compared to the lidocaine group. Conclusion The findings of this meta-analysis back up the claim that articaine is more effective than lidocaine in providing anaesthesia in patients with IP. PROSPERO Registration No.: CRD42020204606 (https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42020204606).


INTRODUCTION
Dental caries is the most widespread non-communicable disease and a major public health concern globally. It is also the most common preventable disease, recognised as the primary cause of oral pain and tooth loss (World Health Organization, 2017). Pulpitis is the inflammation of the dental pulp and a sequel to caries. It is clinically classified as reversible or irreversible (Rôças et al., 2015). Irreversible pulpitis (IP), usually develops when the pulp is exposed to caries biofilm causing necrosis or death of pulp tissues, indicating the need for endodontic treatment (Li et al., 2012;Zanjir et al., 2019).
Successful anaesthesia is a hallmark of painless endodontic treatment. It not only keeps the patient calm and relaxed but also allows the dentist to perform the endodontic procedure with ease (Subbiya & Pradeepkumar, 2016;Howait & Basunbul, 2019). Pulpal anaesthesia commences with the administration of traditional nerve blocks (NBs) or infiltration anaesthesia. For maxillary teeth, NB injections are administered at the anterior superior alveolar (ASA), middle superior alveolar (MSA), and/or posterior superior alveolar (PSA) nerves. For mandibular anterior teeth and molars, both the inferior alveolar nerve block (IANB) and buccal infiltration (BI) injections are administered (Reader, Nusstein & Hargreaves, 2011). Local anaesthetic (LA) drugs have a peripheral effect and block the transmission of nerve impulses. Factors that affect anaesthetic drug efficacy are the type of applied drug, correct injection site, injection velocity, and amount and dosage of the injected drug. Also, the presence of inflammation at the injection site should not be overlooked (Modaresi et al., 2016).
The success rate of achieving deep pulpal anaesthesia lowers in patients with IP. The success rate of IANB can be reduced to <30% (Shahi et al., 2018), and that of maxillary NBs to <60% (Sherman et al., 2008). It is broadly accepted that achieving anaesthesia in patients with IP is more complex, as compared to normal, healthy pulps (Dou et al., 2013;Dou et al., 2018). Inflamed pulp shows lower pH levels, lowering the penetration of basic anaesthetic into the nerve membrane, thus delaying or preventing pulpal anaesthesia (Zanjir et al., 2019). This state of the tooth is frequently referred to as a 'hot pulp', which requires supplementary approaches to ensure a pain-free treatment (Subbiya & Pradeepkumar, 2016). Nonetheless, the effect of different anaesthetic agents and techniques along with or without supplemental infiltration needs to be assessed.
Lidocaine, the first commercialised amide LA solution shows rapid onset when used for most of the dental treatments and is considered as the gold standard LA agent due to its high efficacy, low allergenicity and minimal toxicity (Su et al., 2016). Articaine [(4methyl-3-[1-oxo-2-(propylamino)-propionamido]-2-thiophene-carboxylic acid methyl ester hydrochloride)] is a unique amide LA agent that contains a thiophene ring instead of a benzene ring, demonstrating increased liposolubility and high tissue penetrability as compared to lignocaine. The thiophene ring raises the diffusion of the anaesthetic solution into the cortical bone, thereby penetrating the mandibular dense cortical bone as well as maxillary cortical plates. Previous studies by De Geus et al., (2020) and Srinivasan et al. (2017) have also shown that articaine is equally effective in comparison to other anaesthetics with the success rate ranging from 64 to 87%.
Systematic reviews comparing the anaesthetic efficacy of articaine and lidocaine for dental procedures have been published. Kung, McDonagh & Sedgley (2015) concluded that articaine has a significant advantage over lidocaine as a supplementary infiltration after mandibular block anaesthesia but no advantage when used alone as mandibular block or maxillary infiltration anaesthesia. Su et al. (2016) stated that at the injection phase and treatment phase, 4% articaine is superior in controlling pain and increasing the success rate of local anaesthesia than 2% lidocaine. However, out of 24 articles included in that analysis, 20 articles were Chinese language reports which could not be accessed and retrieved from the databases by the authors of this review. Both the reviews described above were based on searches conducted until 2013. De Geus et al. (2020 in the network meta-analysis concluded that in patients with IP, the use of articaine increased the IANB success rate. This review considered the efficacy of IANB in patients with IP and did not include maxillary injection techniques as well as other supplemental techniques. A preliminary electronic database search revealed that since the publication of the above reviews, several new randomised clinical trials (RCTs) comparing articaine and lidocaine for patients with symptomatic IP have been published. Therefore, the present systematic review assessed the efficacy of 4% articaine vs 2% lidocaine in the mandibular and maxillary block and infiltration anaesthesia in patients with IP.

Protocol development
This systematic review and meta-analysis are written and reported according to the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) statement and registered in PROSPERO (CRD42020204606). The following focused question in the Patient, Intervention, Comparison, and Outcome (PICO) format was proposed ''Is there a difference in the efficacy of 4% articaine vs 2% lidocaine in the mandibular and maxillary block and infiltration anaesthesia in patients with IP?''

Search strategy
To obtain publications in the English language, a complete electronic search was conducted through July 2020 on databases such as PubMed and MEDLINE, Cochrane Central Register of Controlled Trials, and Web of Science. A detailed electronic search of the journals listed in Table 1 was carried out. The searches in the clinical trials database, cross-referencing and Grey literature were conducted using Google Scholar, Greylist, and OpenGrey. Articles were found using Medical Subject Headings (MeSH) terms, keywords, and other free terms coupled with Boolean operators (OR, AND). Following the syntactic guidelines of each database, the same terms were utilised across all search platforms. Table 1 shows the search strategy, population, interventions, comparisons, outcomes, and study design (PICOS) tool.

Population (P):
Studies on patients with symptomatic IP, requiring endodontic treatment under maxillary and mandibular infiltration or blocked anaesthesia on at least one tooth in the mandibular or maxillary region irrespective of age, gender, race, or socioeconomic status, were evaluated. Active responses to spontaneous pain, thermal tester, cold tester, or an electronic pulp tester were considered as diagnostic criteria for IP.
Interventions (I): Studies using 4% articaine as a LA solution for the treatment of IP. Comparison (C): Studies using 2% lidocaine as a LA solution for the treatment of IP. Outcome (O): Primary outcome: It included the success rate assessed as 'no/mild' pain during access cavity preparation and biomechanical preparation phase on the Heft Parker visual analogue scale (VAS).
Secondary outcome: Onset of anaesthesia assessed from the time lapse between the end of the NB and the onset of symptoms of subjective anaesthesia (feeling of heaviness at the site of injection) were calculated in seconds, and the pain was assessed quantitatively during the treatment.

Exclusion criteria
• Studies involving patients with a significant medical history or on medicaments that may affect the anaesthetic assessment.
• Observational study designs, case reports, case series, cross-sectional studies, and reviews.
• Trials reporting a single intervention.
• Article reporting only abstracts and full-texts were not available in the database.

Screening process
Two reviewing authors did the search and screening according to the previously stated methodology (SP & AM). The titles and abstracts were examined first. Second, full-text publications were picked for in-depth reading and analysis based on the data extraction criterion's inclusion and exclusion criteria. Cohen's kappa (k) determined the degree of agreement between the two reviewers to be 0.94 for titles and abstracts and 0.96 for full-texts. After discussions, the third author SM was able to reconcile the disagreements among the authors/reviewers. The authors of the listed papers were contacted via email for clarification of any concerns or missing data.

Data extraction
The two independent authors (IA & BA) extracted the following data from the included studies: author names, study design, tooth, sample size, method of pulp testing, type of local anaesthesia used, injection technique, method of analysis, method of outcome assessment, follow-up and author's conclusions.

Assessments of the risk of bias and quality
The level of evidence for included studies was assessed using the Joanna Briggs Institute (JBI) level of evidence (The Joanna Briggs Institute, 2014). The quality of the selected studies was assessed using the Cochrane Collaboration Tool (Higgins et al., 2019) for RCTs, including random sequence generation, allocation concealment, blinding of participants, incomplete outcome data, selective reporting, and other biases. The quality of NRS was assessed using the Minors checklist (Slim et al., 2003), wherein the minimum outcome assessment time of 5 min was considered appropriate for the included studies.

Statistical analysis
Review Manager (RevMan) 5.3 was used for statistical analysis. The combined results were expressed as relative risks (RRs) for the dichotomous data at 95% confidence intervals (CIs) and P < 0.05 was considered significant. Statistical heterogeneity was assessed by the I 2 test at α = 0.10. Subgroup analysis was conducted for I 2 > 50% and P ≤ 0.10. For I 2 > 50%, the random-effects model was applied. Sensitivity analysis was conducted to assess the stability of the results. Funnel plots were drawn to detect publication bias for studies exceeding 10 in number for each outcome assessed (Su et al., 2016).

Literature search
The PRISMA statement flowchart summarising the selection process is presented in Fig. 1. Among 33 full-text articles, 26 were selected after pre-screening, applying the eligibility criteria, and addressing the PICOS question. Seven studies were excluded as three did not have a control group, three had inappropriate population variables, and one applied intraosseous injection technique, hence only 26 studies were included in the qualitative analysis, whereas 22 out of 26 studies for quantitative synthesis.
The mandibular postoperative pain assessed by Ghazalgoo et al. (2018), Umesh (2017), Zain et al. (2016),  and  using Heft Parker VAS did not show a statistically significant difference between the two groups. The time point for outcome measurement and the outcomes of interest, based on the scoring criteria, were different; also, the method of delivery of the anaesthetic solution varied in each of the included studies. Thus, these were precluded from the meta-analysis, and only qualitative analysis was conducted ( Table 2).
The study by  concluded that 4% articaine was more efficacious than 2% lidocaine, and AMSANB was more beneficial than IONB in achieving the anaesthetic effect of maxillary anterior teeth and premolars. It was a single study included for assessing the postoperative pain using the visual analogue scale and the onset time of anaesthesia for maxillary teeth and was not considered for qualitative synthesis. The onset time of anaesthesia for mandibular teeth was assessed by Tortamano et al. (2013) and concluded that 4% articaine exhibited rapid onset with the highest duration of pulpal anaesthesia in IANB. However, since this was a single study, it was not considered for quantitative synthesis ( Table 2).

The success rate of anaesthesia for maxillary teeth
The pooled outcomes from 3 studies Nabeel, Ahmed & Sikander, 2014; on tooth level unit for maxillary buccal infiltration technique showed that 4% articaine resulted in a success rate of 1.06-fold higher than that for 2% lidocaine (RR, 1.06; 95% CI [0.95-1.2]; P = 0.3, I 2 = 0%) using random-effects model, showing statistically insignificant difference (Fig. 4). Table 4 represents the results of sensitivity analysis for mandibular teeth. Studies of high risk Zain et al., 2016;Nabeel, Ahmed & Sikander, 2014;Tortamano et al., 2013;, moderate risk Monteiro et al., 2015;Sood, Hans & Shetty, 2014) or low risk of bias (Gao & Meng, 2020;Kumar et al., 2020;Aggarwal et al., 2019;Ghazalgoo et al., 2018;Lokhande et al., 2019;Martínez-Martínez, Freyle-Granados & Senior-Carmona, 2018;Shapiro et al., 2018;Umesh, 2017;Monteiro et al., 2015;Ahmad et al., 2014;Nabeel, Ahmed & Sikander, 2014;Sood, Hans & Shetty, 2014;Tortamano et al., 2013;Poorni et al., 2011) were excluded from sensitivity analysis. The clinical success rates comparing the articaine and lidocaine groups showed a significant change after the exclusion of these studies. Reanalysis using the fixed-effect model also showed that the outcomes were not adverse. The exclusion of subgroups with a single study showed a significant improvement in the success rate of articaine group as compared to the lidocaine group in the treatment of IP. Moreover, the subgroup analysis for different injection techniques showed a significant change after the inclusion of only supplementary BI after IANB failure technique as compared to the inclusion of IANB and BI technique (Table 4).

Sensitivity analysis
Publication bias for studies included on mandibular anaesthetic technique was evaluated using a funnel plot (Fig. 5). The funnel plot showed asymmetry at the apex from the centre line having more studies on the right side as compared to left, representing a lack of inclusion of publications describing non-significant intervention results as well as the omission of unpublished studies that might result in an over-estimation of the true effect of an intervention.

DISCUSSION
Endodontic pain management is a critical component in reducing extreme anxiety during endodontic treatment. LA's in-depth and methodical expertise, as well as its suitable delivery methodologies, are essential for pain-free dental treatment . The gold standard amide anaesthetic, lidocaine, has a brief start of action, but when combined with epinephrine, the duration of action increases to intermediate . In patients with endodontic discomfort, the success of lidocaine in IANB and infiltration remains low Aggarwal, Jain & Debipada, 2009;Hargreaves & Keiser, 2002). The risk of failed local anaesthetic was eight times greater in patients with IP than in normal individuals (Hargreaves & Keiser, 2002). Pain is initially transmitted by A-delta and C-fibres in IP, but as the inflammatory process advances, C-fibre transmission takes over, resulting in changed pain characteristics. Strong, quick, acute, and well-localized pain is caused by A-delta fibres, whereas dull, persistent, and radiating pain is caused by C-fibres (Prpic-Mehicic & Nada, 2010). Anatomical causes, acute tachyphylaxis, and the influence of inflammation on local tissue pH, blood flow, nociceptors, central sensitization, and psychological variables are all possible reasons for failure Poorni et al., 2011;Aggarwal, Jain & Debipada, 2009;Reader, Nusstein & Hargreaves, 2011). Articaine is more effective in reducing the action potential produced by A-fibres as compared to 2% lidocaine and the complete disappearance of the action potential produced by C-fibres   (Fig. 6).
This systematic review and meta-analysis of clinical trials provides level 1 evidence for evaluating the efficacy of 4% articaine and 2% lidocaine in the mandibular and maxillary in various countries (24 randomised trials and 2 non-randomised trials). The study participants ranged in age from 15 to 65 years old and were of both genders. Therefore, the findings of this systematic review may be applied to a wide variety of people, as well as anaesthetic effectiveness in mandibular and maxillary teeth with symptomatic IP. Pulp sensitivity tests and the exclusion of periapical diseases were used to determine if research subjects met the criteria for symptomatic IP in the included studies. Participants who have been taking any medication that may affect the impact of local anaesthetic were also excluded from the research, reducing selection bias. The most common method for anaesthetizing mandibular teeth is IANB (Zain et al., 2016). However, IANB is an unreliable anaesthetic method, especially in the case of IP; even when correctly administered, the success rate varied from 15-25% Shapiro et al., 2018;Zain et al., 2016). This emphasises the need of having alternatives to IANB . Thus, in the present systematic review, the pulpal anaesthetic efficacy between articaine and lidocaine were assessed using various infiltration and block techniques. For mandibular teeth IANB, BI, supplementary BI, and intraligamentary injections after IANB failure, BI combined with intraligamentary injection, standard IANB, and long BIs, IANB plus BI and for maxillary teeth via BI, AMSA and IONB were used. The 2011 meta-analysis by Brandt et al. (2011) found that articaine as an infiltrating agent was 3.8-fold more likely to be successful than lidocaine. A recent meta-analysis by Kung, McDonagh & Sedgley (2015) demonstrated that in cases of a failed IANB, supplementary infiltration with 4% articaine was 3.55-fold more successful in achieving profound anaesthetic effect than 2% lidocaine.
Various criteria were applied in this review to assess the first pulpal anaesthetic success, including lip numbness, cold testing, and electric pulp testing. Lip numbness was noted in the majority of the trials, albeit it was a subjective symptom. Only a few studies used a cold test or an electric pulp test followed by lip numbness to validate the first pulpal anaesthetic success, as there was a weak connection between lip anaesthesia, cold testing, and pulpal anaesthesia following IANB for IP mandibular molars Umesh, 2017). Bjorn was the first to link a negative response to the maximal output of electrical pulp stimulation for painless dental treatment (Bjorn, 1946). Dreven et al. (1987) examined the reaction to an electric pulp tester as a measure of pulpal anaesthetic prior to endodontic treatment in teeth with normal pulp, reversible pulpitis, and irreversible pulpitis. In IP, however, a lack of reaction to cold or electric pulp tests does not always imply pulpal anaesthesia (Dreven et al., 1987). This might be because in teeth with IP, the reactions to electric pulp tests and cold testing are linked to rapid and slow silent A-delta fibres, respectively. Therefore, it can be assumed that if the tetrodotoxin-resistant sodium channels appear on deeper nociceptive C fibres, then neither negative nor positive responses to EPT and cold tests indicate the success of anaesthesia as the C fibres might be accountable for the pain response Brandt et al., 2011). Hence, an appropriate alternative is to record the pain response during access cavity opening and pulp extirpation. All the studies included in the systematic review assessed the clinical success of 4% articaine and 2% lidocaine based on pain response during access preparation using VAS.
In this systematic review, the efficacy of anaesthetic solutions was measured by clinical success rate, postoperative discomfort, and local anaesthesia onset time. The meta-analysis of the collected data from 22 studies that satisfied the inclusion criteria was based on the clinical success rate. Furthermore, there was methodological variability in terms of research location, study setting, sample size, number and expertise of investigators performing procedures and diagnosis, volume of LA solution, epinephrine concentration, and marking on the pain scale. This heterogeneity was addressed by using a random-effects model for meta-analysis (Kung, McDonagh & Sedgley, 2015). This meta-analysis found that when used in mandibular and maxillary block and infiltration anaesthesia, 4% articaine, which is a more concentrated LA solution, was more likely to provide anaesthetic success than less concentrated 2% lidocaine anaesthetic solution. Besides, wide CIs were observed in the forest plot analysis, potentially contributing to heterogeneity as shown by I 2 estimates, i.e., 72%, with statistically significant difference favouring the articaine group for mandibular teeth (Fig. 3) as compared to narrower CIs contributing to 0% heterogeneity favouring the articiane group with non-significant difference favouring the articaine group for maxillary teeth (Fig. 4). Similar results were reported by previous meta-analysis that were conducted (Su et al., 2016;De Geus et al., 2020;Kung, McDonagh & Sedgley, 2015;Brandt et al., 2011). Discrepancies in the chemical properties of the molecular structures of the 4 percent articaine and 2 percent lidocaine LA agents might have resulted in clinical differences. The uncharged form of a LA molecule is required for diffusion across the sheaths of lipid neurons and cell membranes, therefore the anaesthetic dissociation constant (pKa) is a crucial number for successful anaesthesia (De Geus et al., 2020). In addition to ionisation, fat solubility and protein-binding properties contribute to the clinical characteristics of LAs whereas, their clinical performance is influenced by the site of injection, concentration of drug and vasoconstrictor, injection volume, and anaesthetic solution's inherent vasodilatory properties (De Geus et al., 2020;Haas, 2002;Moore & Hersh, 2010). A previous study showed that articaine suppresses the compound action potential of the A fibres in the isolated rat sural nerve (Potonik et al., 2006). Also, ionic channels are blocked even in lower concentrations with the thiophene derivative (articaine) as compared to the benzene derivative (lidocaine) (Kolli, Nirmala & Nuvvula, 2017).
The differences in the method of delivery of LA for mandibular teeth for the included trials were one of the major challenges posed by this meta-analysis. In the sub-group analysis, a significant advantage of using articaine over lidocaine for supplementary infiltration after mandibular block anaesthesia over the mandibular block and infiltration anaesthesia alone was observed with heterogeneity of 36% (Fig. 3). Another meta-analysis by Brandt et al. (2011) showed that the pulpal anaesthetic efficacy of articaine was markedly superior to lidocaine when used during infiltration. One study each for supplementary intraligamentary injection, IANB plus long buccal, and IANB with BI was included in the subgroup analysis for mandibular teeth. However, to evaluate whether the final results were dependent on subgroup results of these single studies, a sensitivity analysis was performed via meta-analysis by excluding the above studies in question. This analysis confirmed that although the exclusion of the studies reduced the RRs and heterogeneity, the overall results were unchanged (Table 4).
The secondary objective of this systematic review was to assess the postoperative pain and mean onset time of the two LA solutions. The group receiving 4% articaine, as opposed to 2% lidocaine, experienced less pain as measured by VAS during the injection and treatment phases, which might be due to articaine's 1.5-fold higher potency than lidocaine's (Su et al., 2016). In terms of onset time, 4 percent articaine was shown to be faster than 2 percent lidocaine in pulpal anaesthesia. This phenomenon might be explained by the fact that the onset period of anaesthesia is proportional to the pace of epineural diffusion. This rate is proportional to the percentage of drug in the base form, which is proportional to the pKa; articaine's pKa was lower than lidocaine's (Su et al., 2016).
Intriguingly, when this study was compared to prior English language systematic reviews (Su et al., 2016;Kung, McDonagh & Sedgley, 2015;Brandt et al., 2011;Katyal, 2010), there were some striking similarities and variations in terms of anaesthetic solution and administration techniques. Except for the reviews by Su et al. (2016) and Kung et al. (Brandt et al. 2011), the main difference between this and previous reviews (Su et al., 2016;Kung, McDonagh & Sedgley, 2015;Brandt et al., 2011;Katyal, 2010) is that all subjects in this review were diagnosed with IP, whereas previous reviews consisted of a broad cohort of patients and non-patient volunteers with or without pain (Srinivasan et al., 2017;Kung, McDonagh & Sedgley, 2015). In comparison to the current analysis, which comprised parallel-design clinical trials, earlier reviews by Brandt et al. (2011) and Katyal (2010) used crossover design. In comparison to prior research, the start and end of the search time changed in the current study. Previous studies mostly looked at adverse events, pain, and the onset of local anaesthetic, but this study examined at the total clinical success rate as well as subgroup and sensitivity analyses.
Nevertheless, the present review has some limitations. It was not possible to fully avoid the clinical heterogeneity among the included studies. The sample size of the studies was small, thus lacking statistical power. The reasons for the difference in postoperative pain and the onset time could not be explained because of lack of evidence. Individual tooth type analysis (incisors, canines, premolars, and molars) was not performed, and age and gender were also not taken into consideration in the analysis. However, 15 studies rated good on methodological validity assessment, exhibiting a low risk of bias. The subgroup and sensitivity analyses were performed to rule out the potential reasons for heterogeneity. Thus, it is suggested that in the future, high-quality clinical trials on the outcomes of onset of anaesthesia and pain assessment at various stages of the treatment procedure of IP should be conducted along with the trials assessing the adverse effects of the two solutions at varying concentrations and sites of injection.