Effect of body temperature on the cyclic fatigue resistance of the nickel-titanium endodontic instruments: A systematic review and meta-analysis of in vitro studies.

Aim: The aim of this systematic review was to compare the effect of body temperature (I) on the cyclic fatigue resistance (O) of nickel-titanium (NiTi) endodontic instruments (P) to that of room temperature (C).
Methods: The study was registered in the PROSPERO database (CRD42020204286). A systematic search in PubMed, Scopus, Web of Science, Google Scholar, and OpenGrey was conducted in English until December 31, 2021. In vitro studies comparing the cyclic fatigue resistance of NiTi instruments at the body (35°C ± 2°C) and room temperature (20°C-25°C) were included. Eligible studies were evaluated for risk of bias and meta-analyzed to estimate the effects.
Results: Twenty-one studies out of 347 met the criteria for inclusion. The meta-analysis included six studies (n = 215) with comparative study parameters. The overall effect sizes (5.49; 95% confidence interval [CI]: 4.04-6.93) were significantly different (P < 0.001), indicating that the mean values at room temperature were significantly (P < 0.001) higher. The effect sizes for full rotary motion (standardized mean difference [SMD]: 4.80; 95% CI: 3.04-6.56) and reciprocating motion (SMD: 6.37; 95% CI: 3.63-9.11) were not significantly different (P = 0.346). Heterogeneity was high (I 2 = 94%). Sensitivity analysis revealed that the SMD values were not significantly different (P > 0.05) from the overall effect size, indicating that none of the studies had an effect on the overall effect size. Conclusions: Within the limitation of the study, the cyclic fatigue resistance of heat-treated NiTi endodontic files is significantly reduced at body temperature when compared to room temperature. Cyclic fatigue testing should be conducted at simulated body temperature. Copyright:

INTRODUCTION

Nickel–titanium (NiTi) engine-driven instruments continue to be the mainstay in performing mechanical debridement and shaping during endodontic treatment. It has revolutionized the root canal preparation technique by decreasing operator fatigue, time, and procedural errors associated with manual instrumentation.[1] Despite the increased flexibility, a major concern related to their use is the possibility of intracanal separation. The reported incidence of separation is in the wide range of 0.4%–23%.[2] This is widely attributed to two mechanisms. One is torsional fatigue which occurs when the file's tip is locked inside the canal while the main body or shaft of the file continues to rotate, exceeding the elastic limit.[3] The second is cyclic fatigue, which results from rotation around a curve with the consequence of repeated tension and compression of metal and, finally, work hardening followed by fracture.[4] Cyclic or flexural fatigue accounts for most fractured instruments during clinical use and has been studied extensively.[5] The various factors affecting the cyclic fatigue of a NiTi instrument include the instrument design, type of alloy, radius of curvature, angle of curvature, movement kinematics, and temperature.[3] Environmental or intracanal or body temperature is an important confounding factor that is least studied.[6] It is relevant since the metallurgy of NiTi alloys exhibits different behaviors at room or body temperature.[7] Earlier, most of the fatigue studies were performed at room temperature. Recently, many studies have reported the dramatic effect of body temperature on the cyclic fatigue resistance of NiTi instruments with a reported 300%–500% impact on their lifetime.[891011] Hence, the present systematic review aimed to evaluate the effect of body temperature on the cyclic fatigue resistance of the NiTi endodontic instruments compared to room temperature. The objective was to determine how temperature affects the cyclic fatigue resistance of NiTi instruments and help the clinicians to learn more about the mechanical behavior of NiTi in clinical situations.

METHODS

Protocol and registration

The current review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guidelines.[12] The review protocol was registered a priori in the PROSPERO database (CRD42020204286).

Research question and eligibility criteria

The PICOS acronym was used to devise the following question: What is the effect of body temperature (I) on the cyclic fatigue resistance (O) of NiTi endodontic instruments (P) when compared to room temperature (C) as measured by in vitro studies (S)?

Population: NiTi endodontic instruments

Intervention: Body temperature (35°C ± 2°C)

Comparison: Room temperature (20°C–25°C).

Outcomes: Cyclic fatigue resistance, i.e., the number of cycles to fracture (NCF) or time to fracture (TTF).

Studies: In vitro studies.

Exclusion criteria included studies assessing cyclic fatigue resistance at temperatures other than those indicated above. In addition, abstracts, conference proceedings, reviews, and studies published in languages other than English were not selected. However, those translated into English were included.

Information sources, search strategy, and selection process

A search of the literature was conducted in three databases: PubMed (1964–2021), Scopus (1960–2021), and Web of Science (1980–2021) until December 31, 2021. Google Scholar (first 100 returns) and OpenGrey databases were searched electronically for unpublished manuscripts, research reports, doctoral dissertations, and other gray literature. The electronic search strategy was developed using the most cited descriptors in previous publications on this theme combining Medical Subject Heading terms and text words. For the database, the following terms were combined: “Body temperature,” “Temperature,” “Environmental temperature,” “Fatigue,” “Cyclic fatigue,” “Fatigue resistance,” “Flexural fatigue,” “Fracture resistance,” “Nickel titanium,” “Niti,” “Nitinol,” “Dental instruments,” “instrument,” and “Rotary.” The Boolean operators “AND” and “OR” were applied to combine the terms and create a search strategy. The search strategies for each database are summarized in Supplymentary Table 1. The search was expanded to include reference lists for screened studies and published reviews. The leading endodontic journals, including the Journal of Endodontics, the International Endodontic Journal, and the Australian Endodontic Journal, were manually searched. Duplicate articles were removed from the database using the Covidence tool (Melbourne, Australia). The selection of studies was performed using a two-stage screening process. This was accomplished by two independent reviewers (SS1 and SS2) screening the title and abstract for appropriate studies and reading the full text. The reasons for exclusion are documented in Supplymentary Table 2.[4243444546] In the event of a disagreement, a third reviewer (AL) was consulted.

Supplementary Table 1

Search strategy for each database

Database	Search strategy
PubMed	(((body temperature[MeSH Terms]) OR (“temperature”[MeSH Terms])) OR (Intracanal temperature[Title/Abstract])) OR (Environmental temperature[Title/Abstract]) AND ((((“fatigue”[MeSH Terms]) OR (cyclic fatigue[Title/Abstract])) OR (fatigue resistance[Title/Abstract])) OR (fracture resistance[Title/Abstract])) OR (flexural fatigue[Title/Abstract]) AND (((((((Nickel?titanium[Title/Abstract]) OR (nickel titanium[Title/Abstract])) OR (Ni?Ti[Title/Abstract])) OR (niti[Title/Abstract])) OR (nitinol[Title/Abstract])) OR (rotary[Title/Abstract])) OR (“dental instruments”[MeSH Terms])) OR (instrument[Title/Abstract])
Scopus	(TITLE-ABS-KEY ((“NiTi”) OR (“nickletitanium”) OR (“nitinol”) OR (“rotary”) OR (“instrument”)) AND TITLE-ABS-KEY ((“fatigue”) OR (“cyclicfatigue”) OR (“flexuralfatigue”) OR (“fatigueresistance”) OR (“resistance”)) AND TITLE-ABS-KEY ((“temperature”) OR (“bodytemperature”) OR (“canaltemperature”) OR (“environmentaltemperature”)) AND TITLE-ABS-KEY ((“rootcanal”) OR (“endodontic”) OR (“preparation”))) AND (LIMIT-TO (LANGUAGE,”English”))
Web of science	TS=(body temperature OR temperature OR canal temperature OR environmental temperature) AND (fatigue OR cyclic fatigue OR flexural fatigue OR fatigue resistance OR fracture resistance) AND (nickel titanium OR NiTi OR niti OR nitinol OR rotary OR instrument OR dental instruments) AND (root canal OR endodontic OR preparation) AND LANGUAGE: (English) Indexes=SCI-EXPANDED Timespan=All years

Supplementary Table 2

List and reason of excluded studies after full text reading

Study	Reason
Grande et al. (2017)[8]	Different temperature (+20°C and -20°C)
Shen et al. (2018)[9]	Exact NCF values not available
Arslan et al. (2020)[11]	Different temperature (saline irrigation at +4°C and room temperature)
Elsewify et al. (2020)[42]	No comparative room temperature group
Shen et al. (2012)[43]	Different study setting (fatigue behavior under various medium)
Keskin et al. (2021)[44]	Different study setting (cyclic fatigue resistance of different instruments)
Scott et al. (2019)[45]	Different study setting (cyclic fatigue resistance of reciprocating instruments)
Alghamdi et al. (2020)[46]	Different study setting (effect of 5 different curvature locations on the fatigue resistance)

NCF: Number of cycles to fracture

Data collection and data items

The following data were recorded in an Excel spreadsheet (Microsoft, Redmond, WA, USA) by the two independent reviewers: author, year of publication, instrument name, size, taper, type of alloy (conventional or heat-treated), sample size, motion type (full rotary or reciprocating), testing model (static or dynamic), angle of curvature, radius of curvature, distance from the tip, immersion media, rotational speed, insertion depth, insertion angle, material of artificial canal, inner diameter of the canal, and NCF or TTF at room and body temperatures [Table 1].

Table 1

Characteristics of the included studies

First author	Year	Instrument type	Type of alloy	Sample size	Type of motion	Rotational speed (rpm)	Size and taper	NCF at room temperature	NCF at body temperature
Arias	2019	EdgeSequel Sapphire	Heat treated	40	Rotary	500	20/0.04	252.5 (212-300.7)	82.9 (71.1-96.8)
				40			25/0.04	340.8 (289.1-401.8)	79.2 (69.4-90.3)
				40			30/0.04	264.2 (226.4-308.2)	51.2 (44-59.7)
				40			35/0.04	328.5 (306.7-352)	77.1 (67.7-87.9)
				40			40/0.04	246.8 (224.8-270.9)	56.5 (49-65.3)
		Vortex Blue	Heat treated	40	Rotary	500	20.04	311.7 (270.7-359)	93.9 (88.3-99.9)
				40			25/0.04	341.7 (311.7-374.7)	141.4 (130.9-152.7)
				40			30/0.04	214.4 (200.9-228.7)	97.7 (86.5-110.3)
				40			35/0.04	147.1 (133.2-162.5)	87.5 (81-94.5)
				40			40/0.04	126.4 (110.2-144.9)	80.8 (73.5-88.8)
Arias	2018	HyFlex EDM	Heat treated	40	Rotary	400	25/0.08	725.4 (658.8-798.8)	717.9 (636.8-809.3)
		TRU shape	Heat treated	40	Rotary	300	25/0.06	234.7 (209-263.6)	83.2 (76-91.1)
Cardoso	2019	ProDesign Logic	Heat treated	24	Rotary	350	30/0.05	1265.34±296.40	799.58±349.54
		XP-endo Shaper	Heat treated	24	Rotary	800	30/0.01	4391.11±481.22	1264.44±171.58
		iRaCe	Conventional	24	Rotary	600	30/0.04	301.67±91.73	145.83±88.20
De Vasconcelos	2016	PTU	Conventional	40	Rotary	300	25/0.08	199±41	134±39
		HyFlex CM	Heat treated	40	Rotary	500	25/0.06	2986±412	487±96
		TRU Shape	Heat treated	40	Rotary	300	25/0.06 v	1372±335	201±35
		Vortex Blue	Heat treated	40	Rotary	500	25/0.06	1816±1176	479±139
Generali	2020	Procodile	Conventional	20	Reciprocating	300	25/0.06	186±62*	126±33*
		Reziflow	Conventional	20	Reciprocating	300	25/0.06	222±32*	174±37*
Huang	2017	K3	Conventional	24	Rotary	300	25/0.04	501.54±70.46	413.75±46.37
		K3XF	Heat treated	24	Rotary	500	25/0.04	914.58±215.30	603.47±112.45
		Vortex	Heat treated	24	Rotary	500	25/0.04	1356.94±126.87	1102.78±84.93
Ismail	2020	WaveOne Gold	Heat treated	40	Reciprocating	350	25/0.07	850±130	833±138
		TFA	Heat treated	40	Adaptive	400	25/0.06	728±151	568±63
		PTN X2 files	Heat treated	40	Rotary	300	25/0.06	388±40	347±29
Keleş	2019	Reciproc Blue	Heat treated	60	Reciprocating	300	25/0.08	214.4±108.4*	254.4±62.9* 253.2±77.2*
		Reciproc	Heat treated	60	Reciprocating	300	25/0.08	196.7±55.6* 292.4±81.3*	224.5±50.6* 272.7±55.6*
		WaveOne Gold	Heat treated	60	Reciprocating	350	25/0.07	175.3±60.3* 258.9±47.9*	150.4±31.9* 220.9±42.1*
		WaveOne	Heat treated	60	Reciprocating	350	25/0.08	106.2±36.6* 177.9±46.9*	102±29.1* 107.3±13.1*
La Rosa	2021	F6 SkyTaper	Conventional	60	Rotary	300	25/0.06	250±28* 235±33*	159±31* 157±34*
		One Curve	Heat treated	60	Rotary	300	25/0.06	245±31* 271±34*	186±30* 173±29*
La Rosa	2021	F6 SkyTaper	Conventional	60	Rotary	300	25/0.06	195±30* 171±27*	149±27* 90±31*
		One Curve	Heat treated	60	Rotary	300	25/0.06	175±28* 115±26*	128±26* 105±28*
Plotino	2017	PTU S1	Conventional	30	Rotary	300	18/0.02	515±90.35	380±39.43
		PTG S1	Heat treated	30	Rotary	300	18/0.02	674.17±86.41	629.17±87.25
		PTU.F2	Conventional	30	Rotary	300	25/0.08	228.33±52.45	114.17±47.14
		PTG F2	Heat treated	30	Rotary	300	25/0.08	504.17±94.41	457.5±101.97
Plotino	2018	Reciproc blue	Heat treated	40	Reciprocating	300	25/0.08 v	395±20*	191±51*
		Reciproc	Heat treated	40	Reciprocating	300	25/0.08 v	150±14*	106±35*
Topocuoglo	2020	HyFlex CM	Heat treated	40	Rotary	500	25/0.6	1612.1±357.6 1472.3±275.5	1125.3±304.2 1035.2±289.4
		One Curve	Heat treated	40	Rotary	300	25/0.6	1552.6±361.2 1404±301.4	1373.2±389.6 1256.4±311.7
		ProTaper Next X2	Heat treated	40	Rotary	300	25/0.6	876.5±268.1 713.2±141.4	586±184.2 504.3±163.3
		Edge file	Heat treated	40	Rotary	400	25/0.6	1753.3±402.5	1466.1±388.3 1315.3±359.2
Vieira	2020	Reciproc blue	Heat treated	24	Reciprocating	300	40/0.06	18.06±3.93*	6.73±1.29*
Vieira	2021	Vortex blue	Heat treated	24	Rotary	500	40/0.04	1.899±629.73	1.033±190.07
		Reciproc blue	Heat treated	24	Reciprocating	300	40/0.06	5.419±1179.85	2.019±388.49
		X1 Blue	Heat treated	24	Reciprocating	300	40/0.06	2.974±449.12	1.082±374.19
Dosanjh	2017	ESX Files	Conventional	60	Rotary	500	25/0.04	466	271
		EdgeFileϯ	Heat treated	60	Rotary	500	25/0.04	7243	1675
		Vortex Blueϯ	Heat treated	60	Rotary	500	25/0.04	2062	1233
Alwafazϯ	2018	PTG F2	Heat treated	30	Rotary	300	25/0.08	1239.1±388.2	962.9±276.0
GundogarϮ	2019	Reciproc Blue	Heat treated	30	Reciprocating	300	25/0.08	7914±1266	1349±161
		HyFlex EDM	Heat treated	30	Rotary	500	25/0.08	9847±1378	1812±198
		WaveOne Gold	Heat treated	30	Reciprocating	350	25/0.07	4626±565	1206±148
		TFA	Heat treated	30	Adaptive	400	25/0.08	3067±429	1139±136
KlymusϮ	2019	Reciproc Blue	Heat treated	20	Reciprocating	300	25/0.08	3473±278.9	1521±109.4
		X1 Blue	Heat treated	20	Reciprocating	350	25/0.06	3726±322.0	1647±192.1
		WaveOne Gold	Heat treated	20	Reciprocating	350	25/0.07	1919±265.6	1532.7±182.4
SaeedϮ	2019	HyFlex EDM	Heat treated	20	Rotary	500	25/0.08	4685.34±726.39	3971.25±1012.52
		PTG F2	Heat treated	20	Rotary	300	25/0.08	1959.4±66.08	1027.29±49.78
		2Shape TS 2	Heat treated	20	Rotary	300	25/0.06	416.87±23.55	198.80±25.45
Staffoli	2019	OneShape	Conventional	40	Rotary	300	25/0.06	462±60.3	297.8±58.8
		OneShape new generation	Conventional	40	Rotary	300	25/0.06	473.8±83.4	295±46.5
		One CurveϮ	Heat treated	40	Rotary	300	25/0.06	1513.1±154.6	657.2±104.1
First autdor	Immersion medium	Testing model	Radius of curvature of tde canal	Angle of curvature of tde canal	Distance of center of curvature from instrument tip	Material of tde artificial canal	Insertion angle	Insertion deptd (mm)	Inner diameter of canal (mm)
Arias	Deionized water	Static	3	60	4.5	Stainless steel	-	-	-
Arias	Water	Static	3	60	5	Stainless steel	-	-	-
Cardoso	Water	Static	5	90	5	Stainless steel	-	-	1.5
De Vasconcelos	Water	Static	3	60	4.5	Stainless steel	-	19	-
Generali	Water	Static	5	60	5	Stainless steel	0	16	-
Huang	Water	Static	5	60	-	Zirconium oxide-ceramic	-	19	-
Ismail	Water	Static	2	60	-	Stainless steel	-	19	1.5
Keleş	Water	Static Dynamic Static Dynamic Static Dynamic Static Dynamic	5	60	5	Stainless steel	-	-	1.5
La Rosa	Water	Static	5	60	-	Stainless steel	0 20 0 20	-	-
La Rosa	Water	Static	5	60	-	Stainless steel	0 3 5 3	-	-
Plotino	5% Naocl	Static	5	60	-	Stainless steel	-	-	-
Plotino	Water	Static	5	60	6	Stainless steel	-	-	-
Topocuoglo	Water	Dynamic	5	60	8	-	-	18	-
			2	70	2
			5	60	8
			2	70	2
			5	60	8
			2	70	2
			5	60	8
			2	70	2
Vieira	Water	Dynamic	5	60	-	Stainless steel	-	17	-
Vieira	Water	Static	5	60	5	Stainless steel	-	-	1.5
Dosanjh	Water	Static	5	60	-	Stainless steel	-	-	1.5
AlwafazϮ	Water	Static	5	60	5	Stainless steel	-	19	-
GundogarϮ	Water	Static	5	60	5	Stainless steel	-	-	1.5
KlymusϮ	Water	Static	5	60	5	Stainless steel	-	-	-
SaeedϮ	Water	Static	5	60	-	Stainless steel	-	19	1.5
Staffoli	Water	Static	5	60	5	Stainless steel	-	16	-

*TTF,Ϯ Included in meta-analysis. PTU: ProTaper Universal, PTG: ProTaper Gold, PTN: ProTaper next, TFA: Twisted File Adaptive, NCF: Number of cycles to fracture

Study risk of bias assessment

The methodological quality of the included research was determined using an adaption of a prior systematic review that included in vitro investigations.[13] The domains listed below were used: (1) sample standardization, (2) sample size calculation, (3) sample randomization, (4) single-operator, (5) blinding, (6) testing model standardization, and (7) appropriate statistical analysis. The domains were categorized as “+” to specify a low risk of bias (RoB) and “-” to indicate a high RoB. The articles were classed as having a low RoB if they had six or more domains classified as low (+), a moderate RoB if four or five domains were classified as low, and a high RoB if only three or fewer domains were classified as low. The two reviewers independently assessed the quality (SS1 and SS2). In a disagreement, a third reviewer (AL) was consulted.

RESULTS

Study selection

Figure 1 presents a flowchart of the systematic review process. A total of 21 studies met the criteria for inclusion.[141516171819202122232425262728293031323334]

Figure 1

The PRISMA flow diagram. PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses

Characteristics of instrument types

The studies examined a total of 29 instruments, nine of which were made of conventional NiTi alloy and eight of which were reciprocating systems. The size of the instrument tip ranged from #20 to #40; however, the majority of studies utilized size #25. In addition, the taper and speed of rotation varied between studies [Table 1].

Characteristics of study design

The studies revealed differences in the test model type, radius of curvature, angle of curvature, and immersion media.[141516171819202122232425262728293031323334] The distance of the curvature from the tip, the fit of the instrument, angle of insertion, and length of the instrument inserted were not mentioned in all the studies [Table 1].

Risk of bias in studies

Table 2 contains a detailed information addressing the RoB in the selected studies. The RoB was evaluated as moderate to high overall.

Table 2

Risk of bias assessment

First author	Year	Sample standardization	Sample size calculation	Sample randomization	Single operator	Blinding	Standardization of testing model	Statistical test	Grading
Arias et al.[14]	2019	−	−	−	−	−	+	+	High
Arias et al.[15]	2018	−	−	−	−	−	+	+	High
Cardoso et al.[16]	2019	+	−	−	+	−	+	+	Moderate
De Vasconcelos et al.[17]	2016	−	−	−	−	−	+	+	High
Generali et al.[18]	2020	+	+	−	−	−	+	+	Moderate
Huang et al.[19]	2017	−	−	−	−	−	+	+	High
Ismail et al.[20]	2020	+	+	−	−	−	+	+	Moderate
Keleş et al.[21]	2019	+	+	+	−	−	+	+	Moderate
La Rosa et al.[22]	2021	+	+	−	−	−	+	+	Moderate
La Rosa et al.[23]	2021	+	+	−	−	−	+	+	Moderate
Plotino et al.[24]	2017	+	−	−	−	−	+	+	High
Plotino et al.[25]	2018	+	−	−	−	−	+	+	High
Topçuoğlu et al.[26]	2020	+	+	−	−	−	+	+	Moderate
Vieira et al.[27]	2020	+	−	+	+	−	+	+	Moderate
Vieira et al.[28]	2021	+	−	+	+	−	+	+	Moderate
Dosanjh et al.[29]Ϯ	2017	−	−	+	−	−	+	+	High
Alfawaz et al.[30]Ϯ	2018	+	−	−	+	−	+	+	Moderate
Gündoğar et al.[31]Ϯ	2019	+	−	+	−	−	+	+	Moderate
Klymus et al.[32]Ϯ	2019	+	+	−	−	−	+	+	Moderate
Saeed and Rafea[33]Ϯ	2019	+	−	−	−	−	+	+	High
Staffoli et al.[34]Ϯ	2019	+	+	−	−	−	+	+	Moderate

Ϯ Included in meta-analysis

Meta-analysis

Because of the heterogeneity among the study design and instrument types, it was decided not to perform a meta-analysis on overall data. To ensure homogeneity, the meta-analysis included only those studies that matched the following criteria:

Heat-treated files

Tip size of 25

Static stainless steel model

Angle of curvature = 60°

Radius of curvature = 5 mm

Water as an immersion media

Cyclic fatigue expressed in NCF.

Six studies were included,[293031323334] examining 10 different instrument types, including the ProTaper Gold (Dentsply Sirona, Charlotte, NC, USA), EdgeFile (EdgeEndo, Albuquerque, NM), Vortex Blue (Dentsply Sirona, Charlotte, NC, USA), Reciproc Blue (VDW, Munich, Germany), HyFlex EDM (Coltene, Cuyahoga Falls, OH, USA), WaveOne Gold (Dentsply Sirona, Charlotte, NC, USA), Twisted File Adaptive (TFA) (SybronEndo, CA, USA), X1 Blue (MK Life, Porto Alegre, RS, Brazil), 2Shape (MicroMega, Besancon, France), and One Curve (MicroMega, Besancon, France). Three of these systems were reciprocating while one had an adaptive (rotation–reciprocating) motion. The TFA changes to a reciprocating mode when engaging dentin or stress. As a result, it was considered under reciprocating subgroup during the meta-analysis [Table 1].

Statistical analysis

STATA version 16.0 (Stata Corp, College Station, Texas, 77845, USA) software was used to carry the meta-analysis to assess whether the mean differences at 20°C and 37°C across the studies were statistically significant. The standardized mean difference (SMD) was calculated using Hedges' g bias correction and was taken as effect size with 95% confidence interval (CI). The fixed-effect model using the inverse-variance method and the random-effect model using the restricted maximum likelihood method were estimated. The heterogeneity was tested across the studies using the I2-statistics using DerSimonian–Laird estimator for tau2. I2-statistics of >50% was considered as significant heterogeneity. The publication bias was assessed using the funnel plot and Begg–Egger regression test. Since two motions (rotary and reciprocating) were adopted, a subgroup analysis was carried out. To assess the consistency of the results, sensitivity analysis was performed by the method of leaving one out study. A meta-regression of SMD on the other study variables available was performed to find out significant contributing factors.

Results of the meta-analysis

Fourteen groups (8 groups with rotary motion and 6 groups with reciprocating motion) were evaluated in six studies with two arms (20°–25°C and 35 ± 2°C). A total of 215 instruments per arm consisting of 140 with rotary motion and 75 with reciprocating motion were studied. All the 14 groups demonstrated that the effect sizes were significantly different, indicating that the SMD was significantly (P < 0.001) higher in the 20°C arm [Figure 1]. The overall effect size for the fixed-effect model was 2.99 (95% CI: 2.66–3.33) and for the random-effect model was 5.49 (95% CI: 4.04–6.93) [Figure 2].

Figure 2

Forest plot analysis comparing the NCF values at room and body temperatures. The subgroups represent different motion kinematics (Motion code 0 = Rotary, Motion code 1 = Reciprocating). NCF: Number of cycles to fracture

Heterogeneity (I2-statistics = 94%) was very high, and it was highly significant (P < 0.001) between the studies. Since the heterogeneity was more than the threshold level (50%), subsequent analyses were restricted to the random-effect model. Subgroup analysis by motion showed that the overall effect size for full rotary motion (SMD: 4.80; 95% CI: 3.04–6.56) and reciprocating motion (SMD: 6.37; 95% CI: 3.63–9.11) did not differ significantly (P = 0.346) [Figure 2]. Sensitivity analysis by leaving out one study was carried out to determine the influence of any particular study on the outcome [Supplymentary Table 3]. The SMD values did not differ significantly (P > 0.05) from the overall effect size by leaving one particular study, confirming that none of the studies influenced the overall effect size.

Supplementary Table 3

Sensitivity analysis

Study omitted	SMD	95% confidence limits		Percentage weight
		Lower	Upper
Alfawaz et al., 2018	5.91	4.40	7.42	91.5
Dosanjh et al., 2017	5.72	4.09	7.36	77.9
Dosanjh et al., 2017A	5.83	4.14	7.52	72.9
Gundogar et al., 2019	5.34	3.87	6.81	96.5
Gundogar et al., 2019A	5.26	3.81	6.71	99.1
Gundogar et al., 2019B	5.27	3.82	6.72	98.8
Gundogar et al., 2019C	5.45	3.95	6.95	92.9
Klymus et al., 2019	5.25	3.80	6.71	98.3
Klymus et al., 2019A	5.33	3.86	6.80	96.2
Klymus et al., 2019B	5.87	4.29	7.44	84.28
Saeed et al., 2019	5.92	4.39	7.44	89.5
Saeed et al., 2019A	5.06	3.66	6.47	99.8
Saeed et al., 2019B	5.27	3.81	6.73	97.8
Staffoli et al., 2019C	5.40	3.92	6.87	95.2
Overall effect	5.49	4.04	6.93	100

SMD: Standardized mean difference

Funnel plot analysis indicated the presence of high publication bias as evident by an asymmetric pattern of the effective size. Further, the intercept of the Begg–Egger regression was highly significant (P < 0.001), confirming the presence of publication bias [Supplymentary Figure 1]. To assess the possible influencing factors for the high heterogeneity level, a meta-regression analysis was carried out by considering the taper size as the covariate. The testing of the regression coefficient of taper size was not statistically significant (P = 0.461), implying that the taper size was not a significant influencing factor toward the high heterogeneity level.

DISCUSSION

The relative proportion and characteristic of the microstructural phases in a NiTi instrument determine its mechanical behavior.[35] The alloy can be classified into two distinct temperature-dependent crystallographic phases: martensite (low-temperature phase) and austenite (high temperature or parent phase), each with its own distinct set of properties.[7] When heated, martensite NiTi transforms into austenite. The austenite start temperature is when this phenomenon begins (As). The temperature at which it is complete is referred to as the austenite finish temperature (Af). When austenite NiTi is cooled to a specific temperature, it transforms into martensite. Similarly, martensite start temperature (Ms) and martensite finish temperature (Mf) exist.[36] The transformation temperatures have a substantial effect on the mechanical characteristics and behavior of NiTi, which can be varied during the production process by minor compositional changes, impurity additions, and heat treatments.[37] The Af temperature of the conventional NiTi files is near or below room temperature. In contrast, the Af temperature of the vast majority of heat-treated files is clearly above body temperature. CM wire, M-wire, and conventional SE NiTi wire have an Af of approximately 55°C, 50°C, and 16°–31°C, respectively.[36] During root canal preparation, the average intracanal temperature is 35.1°C, comparable to body temperature.[38] Thus, conventional NiTi files are predominantly in the austenite phase at or below intracanal temperature, whereas heat-treated files are predominantly in the martensite/R-phase/hybrid phase at intracanal temperature.

The bulk material properties primarily determine fatigue life. A hybrid (austenite–martensite) microstructure containing a trace of martensite is more likely to be fatigue resistant than a completely austenitic microstructure. This is often explained by martensite's stronger resistance to fatigue crack growth than stable austenite. The fatigue crack propagation speed of austenitic structures is significantly faster than that of martensite structures at the same stress intensity level. In addition, due to the energy absorption properties of its twinned phase structure, the martensitic phase transformation exhibits exceptional damping characteristics.[39]

In cyclic fatigue studies, environmental temperature is a crucial confounding variable. Numerous investigations have established a considerable effect of ambient temperature on the cyclic fatigue resistance of NiTi endodontic instruments since NiTi alloys behave differently depending on their metallurgical properties.[891011] Thus, compared to room temperature, the current systematic review studied the influence of body temperature on the cyclic fatigue resistance of NiTi devices. The studies that examined the effect of various temperatures (−20°C to 60°C) were excluded, as cooled or heated irrigant solutions rapidly equilibrate to body temperature inside the root canal.[6] The included 21 studies demonstrated notable heterogeneity in the test model type, curvature radius, angle of curvature, immersion media, and instrument features. The present review demonstrates the importance of developing an international standard for validating a device for cyclic fatigue testing of NiTi rotary endodontic instruments. In an ideal scenario, it would enable the testing of all instruments with a precise trajectory in terms of radius and angle of curvature, fit, and angle of insertion, among other characteristics, allowing the comparison of different instruments.[4]

Six studies were selected with similar study designs, instrument characteristics, and outcome measures to achieve homogeneity.[293031323334] A meta-analysis was performed, which discovered that the cyclic fatigue resistance of heat-treated NiTi endodontic instruments is significantly reduced at body temperature. This can be attributed to the alloy transitioning to the austenitic phase in heat-treated instruments. At room temperature, the various heat-treated files are martensitic and transform to a more austenitic state at body temperature, resulting in a mixed martensitic, R-phase, and austenitic structure. The martensitic to austenitic conversion is not complete at body temperature, but a considerable proportion is already austenitic. However, the crystal lattice of conventional NiTi instruments is almost identical at room and body temperatures; nevertheless, literature reports that fatigue resistance is reduced, albeit slightly.[17] This implies that additional unidentified factors may play a role that warrants further investigation. In addition to alloy, the cyclic fatigue of an endodontic instrument is also affected by the instrument's working kinematics (rotary and reciprocating) and diameter.[4041] Hence, a subgroup analysis was conducted using the motion type. Even though the effect size for the reciprocating motion was higher than the rotary motion, the difference was not statistically significant.

The primary limitation of this meta-analysis was the high degree of heterogeneity attributed to the different instrument brands and designs. The instruments all had the same tip size (#25). However, the tapers ranged from 2% to 8%. Previous research has demonstrated that instruments with a narrower taper exhibit greater cyclic fatigue resistance.[32] As a result, the taper size was tested as a covariate in a meta-regression analysis and found to have no significant effect on the effect size, implying that other variables may have contributed to the high heterogeneity. Each of the heat-treated alloys included in the meta-analysis has a different phase transformation temperature, contributing to the heterogeneity. Given that temperature is the focus of the included studies, the methods used to maintain the temperature during testing varied widely, including using a thermocouple or a hotplate and ice. Again, various study design parameters such as the distance of curvature from the tip, the instrument's fit, and the angle of insertion were not mentioned in all studies and could not be retrieved. In addition, most studies demonstrated a high or moderate RoB. Sensitivity analysis was used to determine the effect of omitting one study on the outcome. However, none of the studies affected the overall effect size. High publication bias was also a limitation, indicating that there may be a bias toward publishing only positive effects or due to language bias.

The current review's strength was that, despite discovering significant variation in cyclic fatigue testing models, it attempted a meta-analysis by including studies with comparable parameters. In this regard, this is the first systematic review to investigate the effect of body temperature on the cyclic fatigue resistance of NiTi endodontic instruments.

CONCLUSIONS

Within the limitations of this systematic review, the overall effect size was significantly higher at room temperature, indicating that the cycle fatigue resistance of heat treated NiTi instruments decreases significantly at body temperature compared to room temperature. As a result, future cyclic fatigue testing should be performed at a simulated body temperature that resembles the intracanal environment.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Funnel plot analysis revealing publication bias