Abstract

Aims

This mini systematic review seeks to analyse the available literature and determine if a 4% articaine solution poses a greater risk of inferior alveolar and/or lingual nerve damage compared to that of 2% lidocaine, when administered for an inferior alveolar nerve block.

Results

After a mini systematic review of the published literature, seven suitable studies were identified: one double-blind random controlled trial (DBRCT) and six retrospective cohort studies. The DBRCT and two of the cohort studies concluded that 4% articaine poses no greater risk of nerve damage. The remaining four cohort studies suggested that caution should be exhibited when using a 4% local anaesthetic solution rather than a 2% solution. However, these studies also concluded that no evidence exists to explain the reasons for their results.

Discussion and conclusion

The included articles present no conclusive evidence to suggest that 4% articaine causes more nerve damage than 2% lidocaine, although some authors advise caution when using this agent. All studies conclude that further quality research is required, and it is therefore suggested that dental practitioners exhibit caution when choosing to use 4% articaine in an inferior alveolar nerve block until further scientific research has been performed.

Introduction

Since 1949, lidocaine has been recognised as the 'gold-standard' of local anaesthetic (LA) agents.1 However, the desire to develop fast-acting agents with a short half-life that also produce profound anaesthesia has led to the development of other alternatives. One example is articaine, initially synthesised in 1969 and used for the first time in clinical dental practice in Germany in 1976.

The reason for articaine's popularity appears to be due to its efficacy. Numerous studies have shown that articaine produces a more profound anaesthesia than that of lidocaine.2,3,4,5,6,7,8 Lidocaine is an amide compound, based on a benzene ring structure (C6H6). Articaine, in contrast, possesses a thiophene ring (C4H4S), providing greater lipid solubility and an increased potency as a greater volume of an administered dose can enter the target neurons. Articaine's lipid solubility has been quoted at over four times greater than that of lidocaine.9 The same study confirmed that the onset of anaesthesia was achieved in 7.4 mins with articaine, as opposed to 8.7 mins with lidocaine.9 It has also been suggested that articaine provides a longer duration of anaesthesia due to its protein binding characteristics.10,11

With these attributes, it is perhaps not surprising that many studies have concluded that articaine is more efficient at producing profound anaesthesia than lidocaine.6,12,13,14,15 These papers include studies of both infiltration and nerve block anaesthesia. Other authors concluded that articaine has a faster onset than lidocaine,11 and a meta-analysis has proved that articaine is 1.6–3.5 times more potent than lidocaine.2 Several studies have concluded that articaine should be recommended for use over lidocaine.2,6,12,16 In 2007, Robertson et al. concluded that both the speed of onset and the anaesthetic efficacy of articaine were superior to those of lidocaine, when administered via a buccal infiltration technique in the posterior molar region.14

Another important attribute of a local anaesthetic agent is that of safety and this is perhaps where articaine compares less favourably. Since its introduction, several articles have been published warning of possible nerve damage when articaine is administered in an inferior alveolar nerve block (IANB).17,18 These articles indicate a risk of causing temporary or permanent paraesthesia of the inferior alveolar nerve (IAN) but evidence also exists contradicting these claims.3,19,20

It appears, therefore, that the dental profession faces a dilemma. Should the more efficient agent be used to achieve faster, more profound anaesthesia; or should the profession be wary of an agent that may have the potential to induce nerve damage?

A mini systematic review of the literature was performed by a single researcher with one, clearly focused question.21 The results of the study will hopefully provide advice to the dental profession, ensuring the continued provision of safe and effective local anaesthesia.

Methodology

The Scottish Intercollegiate Guidelines Network (SIGN) presents eight levels of evidence-based research. The SIGN tool was used in this study according to the criteria set out in the hierarchy of evidence.22 The development of the research question was aided using the PICOS method,23 as described in Table 1. Inclusion and exclusion criteria were applied to the literature search as outlined in Tables 2 and 3. Basic search terms and medical sub-headings terms were developed and detailed in Boxes 1 and 2. Three electronic databases were chosen to systematically search the available literature (Table 4):

  1. 1

    MEDLINE with Full Text

  2. 2

    Dentistry & Oral Sciences Source

  3. 3

    The Cochrane Library.

Table 1 PICOS parameters applied to the study
Full size table
Table 2 Search inclusion criteria
Full size table
Table 3 Search exclusion criteria
Full size table
Table 4 Search strategy, 18 November 2016
Full size table

Quality assessment of studies

To ensure that the random controlled trials included in the review were accurately assessed against the inclusion and exclusion criteria, the risk of bias tool as described in the Cochrane Handbook for Systematic Reviews of Intervention was applied.24

For the selected cohort studies, a methodology index for non-randomised studies (MINORS) was applied,25 as described in Table 5. A record sheet was developed, and each study was subsequently scored as directed by Slim and Nini et al. 200325 as defined in Table 6.

Table 5 Methodology index for non randomised studies (MINORS)25
Full size table
Table 6 MINORS criteria scores
Full size table

Data extraction

Specifically designed data extraction forms were developed, allowing uniform data to be extracted under the following headings:

  • Study design

  • Study objectives

  • Geographical origin of the study

  • Clinical setting for the study

  • Study funding

  • Study participants – sex, age, numbers

  • Type of anaesthetic agent used

  • Study outcome–methods of recording and reporting nerve damage

  • Comparison made between 'expected' and 'observed' outcomes

  • Follow-up periods

  • Attrition bias

  • Data analysis of outcomes.

Results

Data extraction and results of the mini systematic review are detailed in Tables 7, 8, 9, 10, 11, 12a, 12b, 13, 14a, 14b and Figure 1.

Table 7 Search strategy and results (performed on 30 December 2016)
Full size table
Table 8 Included studies
Full size table
Table 9 Examples of excluded studies
Full size table
Table 10 MINORS checklist for included studies
Full size table
Table 11 Risk of assessment bias24
Full size table
Table 12a Data extraction
Full size table
Table 12b Data extraction
Full size table
Table 13 Summary of outcome characteristics of included studies
Full size table
Table 14a Summary of study findings
Full size table
Table 14b Summary of study findings
Full size table
Figure 1
figure 1

PRISMA diagram indicating selection/inclusion process

Full size image

Discussion

Malamed and Gagnon's study of 1,325 participants enabled a statistical analysis of the results which indicated that the incidence of nerve damage was the same (1%) whether 4% articaine or 2% lidocaine was used as the LA agent. Indeed, this DBRCT concluded that articaine is a 'safe and effective' local anaesthetic agent.19

Both studies conducted by Pogrel,20,26 concluded that the incidence of nerve damage following the use of 4% articaine was in proportion to its market share. However, three of the studies indicated that the use of 4% articaine elicited more adverse outcomes than would be expected when compared to the agent's market share.17,27,28

Limitations and characteristics of included studies

Several methodological inconsistencies exist throughout the included studies, making a direct comparison between the chosen articles difficult. When performing a study comparing two pharmaceutical agents, a true comparison can only be achieved with the knowledge of the relative use of the two drugs within the studied population. Haas and Lennon,17 Gaffen and Haas,28 and Garisto, Gaffen et al.,27 all used the 'null hypothesis' developed by Ronald Fisher.29 However, the other included studies failed to indicate any comparison between expected and observed outcome events.

The creation of a 'barb' on the tip of the needle, resulting from contact with bone, may also be a factor in the traumatic damage to both the IAN and lingual nerve (LN). However, whether or not this event occurred during any of the IANBs included in the studies, the resultant mechanical damage would be the same for both LA solutions.

Of the seven included papers, only one involves a DBRCT, three involve voluntary reporting of nerve damage, and the remaining three articles elicit their information from patients who have been referred to a specialist centre for the specific reason that they are experiencing some degree of nerve damage. This clearly results in a considerable degree of reporting bias.

With incidences of nerve damage ranging from 1:27,000 to 1:785,000,17,30 it is clear that this study's outcome is extremely rare. To obtain statistically significant results in a DBRCT would require a clinical trial on a very large scale. This could explain the existence of only one such study since 1976.19

Both Hillerup and Jensen,18 and Garisto and Gaffen,27 make reference to the possibility of reporting bias in their papers, and Gaffen and Haas28 admit that 'reported incidence numbers should be viewed cautiously.' In his 2007 paper, Pogrel26 states that he estimates his study represents approximately 10% of all cases of nerve damage in the given population per year. However, reporting bias for patients referred to a specialist centre would be the same for both LA solutions.

The only study that included a detailed physical examination of the patient was that of Hillerup and Jensen,18 using a 'standardised test of neurosensory functions' by a single operator to determine the presence and extent of any reported nerve damage.31,32 The remaining included studies merely noted the incidence of 'reported' nerve damage.

Pogrel's studies,20,26 using data from a specialist centre and Garisto and Gaffen's paper,27 all failed to accurately examine the patient, relying instead on the patient's own descriptions and a log of reported cases to the adverse event reporting system (AERS). Pogrel's description of the patient 'examination' lacks sufficient detail to allow exclusion of detection bias.

The description of the reporting of an 'electric shock' during the administration of the LA created notable discussion among the included authors. Four of the included papers noted the reporting of this phenomenon17,18,27,28 and all included these reports in their results as a 'nerve injury'. The remaining three papers failed to mention this possible event.19,20,26

Interestingly, Hillerup and Jensen state that 'electric shock per se is probably of minor relevance for the aetiology of injection injuries.'18 However, they then go on to question the cause of nerve injury, admitting that it is unknown as to whether the nerve is damaged via neurotoxicity or mechanically, via intra-fascicular injection.

Many authors are now advocating the use of 4% articaine in infiltration anaesthesia as an alternative to block anaesthesia due to the increased efficacy of this agent.33,34,35,36 The evidence presented in these studies indicates a clear efficacy advantage when using 4% articaine as a buccal infiltration compared to 2% lidocaine in an IANB. One author has even suggested that the IANB may now be an unnecessary procedure.37

Concentration of the LA agent

Three of the chosen papers postulate that it may be the fact that, as articaine is administered in a 4% solution, it is the concentration of the LA solution rather than the actual pharmacology of the agent that causes damage to the nerve.17,27,28 This suggestion would appear to be confirmed by another study on rat sciatic nerves, which concluded that significantly more neurotoxic injuries were observed following the direct injection into the nerve of a 4% articaine solution compared to that of a 2% solution.38

However, in a recent in vitro study, articaine proved to be less neurotoxic than lidocaine, mepivacaine and prilocaine.39 Indeed, previous studies have concluded that no scientific evidence exists to confirm the suggestion that articaine causes increased paraesthesia and, to date, no causal relationship has been exhibited between an anaesthetic agent's concentration and neurological damage.40,41

Implications for clinical research

This mini systematic review confirms that controversy still exists over the safety of 4% articaine and 1:100,000 adrenaline as a dental local anaesthetic agent.

The authors of all the included papers admit that, due to the extremely rare occurrence of the outcome, a carefully performed, high quality DBRCT would have to involve such vast numbers of participants that, logistically, such a study would pose certain problems.

It is generally accepted that 4% articaine exhibits greater lipid solubility, faster onset and increased duration of anaesthesia, more profound anaesthesia, and reduced toxicity than those of its counterpart, 2% lidocaine. With these favourable attributes, 4% articaine does indeed offer superior properties over 2% lidocaine but would a 2% articaine solution offer the same advantages?

Further research is required into the efficacy and safety of a 2% articaine solution. Indeed, a study in 2006 proved that the 4% articaine solution was not superior in its anaesthetic effect compared to 2% and 3% solutions of the same agent.42

Implications for general dental practice

The highest level of evidence available to this study was that of Malamed and Gagnon's DBRCT in 2001.19 Although spread over 27 sites in two countries, this trial unfortunately exhibited several potential areas of bias. It did, however, conclude that there was no evidence to suggest that 4% articaine posed a greater risk of nerve damage than 2% lidocaine and that the use of 4% articaine in general dental practice can, therefore, be deemed safe and efficient.

Three further papers, not included in this study, also concluded that no conclusive evidence exists to suggest that 4% articaine poses a greater risk of nerve damage compared to other LA agents.3,10,12

Conclusion

This mini systematic review of the literature has highlighted the fact that further research is required to determine the relative risks of using 4% articaine compared to 2% lidocaine in IANBs. Clearly, the use of 4% articaine is becoming increasingly popular as a means of achieving successful dental anaesthesia and, if current trends continue, this agent may become the number one anaesthetic of choice in the future. This steady increase in popularity is likely to be due to the proven efficacy of this LA agent, benefiting both the patient and the operator. Indeed, the incidence of inferior alveolar nerve damage may reduce in the future as more evidence emerges to support infiltration anaesthesia. With this in mind, and considering the contradictory evidence presented in this study, it is suggested that until factual evidence becomes available, dental practitioners should consider all the potential risks and benefits of a particular LA agent prior to its administration.