Clinical Research

Evaluation of Accuracy, Reliability, and Repeatabilityof Five Dental Pulp TestsEugene Chen, BDSc, BScDent(Hons), DClinDent, and Paul V. Abbott, BDSc, MDS

Abstract

Introduction: The aim of this study was to compare theclinical accuracy, reliability, and repeatability of laserDoppler flowmetry (LDF), an electric pulp test (EPT),and various thermal pulp sensibility tests. Methods:Pulp tests were done on 121 teeth in 20 subjects byusing LDF, EPT, and thermal pulp testing (CO2, EndoFrost [EF], Ice) during 2 or 3 test sessions with at least1-week intervals. The order of testing was reversed onthe second visit. A laser Doppler flowmeter was usedto measure mean pulp blood flow (Flux) calibratedagainst a brownian motion medium and zeroed againsta static reflector. The laser source was 780 nm, with0.5-mm fiber separation in the probe, 3.1 kHz as theprimary bandwidth for filter set to 0.1-second timeoutput constant. Customized polyvinylsiloxane splintswere fabricated for each participant, and a minimumof 90-second recording time was used for each tooth.Raw data were analyzed by using repeated measureanalysis of variance, pairwise comparisons, and inter-class correlations (ICC). Results: The accuracy of EPT,CO2, and LDF tests was 97.7%, 97.0%, and 96.3%,respectively, without significant differences (P > .3).Accuracy of EF and Ice was 90.7% and 84.8%, respec-tively. EPT (P = .015) and CO2 (P = .022) were signifi-cantly more accurate than EF. LDF was more accuratethan EF, but this was not statistically significant(P = .063). Ice was significantly less accurate than EPT(P = .004), CO2 (P = .005), LDF (P = .006), and EF(P = .019). With the exception of Ice (effect of visit:F2,38 = 5.67, mean squared error = 0.01, P = .007,h2p = 0.23), all tests were reliable. Ice (ICC = 0.677)and LDF (ICC = 0.654) were the most repeatable ofthe tests, whereas EPT (ICC = 0.434) and CO2 (ICC =0.432) were less repeatable. Conclusions: CO2, EPT,and LDF were reliable and the most accurate tests, butCO2 and EPT were less repeatable yet less time-consuming than LDF. EF was reliable but not as accurateas EPT and CO2 and less repeatable than Ice and LDF. Icewas the most repeatable but the least accurate and leastreliable test. (J Endod 2011;37:1619–1623)

From the School of Dentistry, University of Western Australia, NAddress requests for reprints to Professor Paul V. Abbott, Schoo

E-mail address: paul.v.abbott@uwa.edu.au0099-2399/$ - see front matter

Copyright ª 2011 American Association of Endodontists.doi:10.1016/j.joen.2011.07.004

JOE — Volume 37, Number 12, December 2011

Key WordsElectric pulp tests, laser Doppler flowmetry, pulp sensibility tests, thermal pulp tests,vitality tests

Cold tests and electric pulp tests (EPTs) are pulp sensibility tests that have been es-tablished as useful aids in the assessment of pulp status (1–3), despite being

subjective and patient-dependent (4, 5). Nonetheless, the lack of pulp sensibility isoften associated with advanced pulp necrosis (6).

Pulp sensibility tests have limitations, and false responses can occur (7). EPTs areknown to be unreliable in immature teeth (3, 8) and in teeth undergoing orthodonticmovement (3). Furthermore, false responses are known to occur when the electriccurrent is conducted to adjacent periodontal tissues, adjacent teeth (9), and even toremnant inflamed pulp tissue with liquefactive pulp necrosis (2). In traumatized teeth,EPTs are less reliable than cold tests (7, 10).

Cold tests are known to elicit false-negative responses in elderly patients becauseof the amount of thermal insulation provided by secondary dentin (2). Commonly avail-able agents for cold tests are CO2 snow, ice, and refrigerant sprays (eg, tetrafluoro-ethane, butane, propane, isobutane, dicholorofluoromethane [DDM], and ethylchloride).

Ethyl chloride was popular (11), but CO2 snow and refrigerant such as DDM havebeen proved effective and even superior (6, 12) than ice and ethyl chloride. DDM hasdecreased in popularity because of environmental concerns (13), and it has been re-placed by other gases such as the mixture of propane, butane, and isobutane (EndoFrost [EF]; Roeko, Langenau, Germany). Concerns have been raised about possibledamaging effects of cold testing agents to teeth, but studies have shown that they aresafe (14–18).

Research into the application of laser Doppler flowmetry (LDF) in traumatizedpulps has been extensive (19–22). The general aim has been to objectively measurethe true vitality of the pulp, that is, the blood supply rather than the sensory function.

There have been few direct comparisons reported between LDF and othermethodsof pulp testing. Ing�olfsson et al (23) compared LDF with EPT, and Evans et al (24)compared LDF with ethyl chloride and EPT. Both studies showed promising resultsfor LDF.

The present study was designed to compare the clinical accuracy, reliability, andrepeatability of LDF and various traditional pulp sensibility tests.

Materials and MethodsTwenty healthy patients (12male and 8 female) participated in the study at the Oral

Health Centre of Western Australia. The mean age was 30 years, with a range of 18–74years. The Human Research Ethics Committee at the University of Western Australiaapproved the study protocol. All subjects were tested 3 times. An interval of at least 1

edlands, Western Australia, Australia.l of Dentistry, University of Western Australia, 17 Monash Avenue, Nedlands, WA 6009, Australia.

Accuracy, Reliability, and Repeatability of Pulp Tests 1619

TABLE 1. Morphotypes of the Teeth Tested

Teeth morphotype No. of teeth

Maxillary central incisors 29Maxillary lateral incisors 22Maxillary canine 12Maxillary premolars 24Maxillary molars 12Mandibular incisors 8Mandibular canine and premolars 8Mandibular molars 6Total 121

Clinical Research

week was left between each test session. All examinations and tests werecarried out by 1 investigator for standardization purposes.

In the first session, all teeth had provisional diagnosis of the pulpstatus made through clinical and radiographic examinations beforepulp testing the teeth. Pulp diseases were confirmed by subsequentendodontic treatment carried out by the relevant practitioners chosenby the test subjects. Each subject had up to 11 teeth selected for pulptesting, with a total of 121 teeth tested. The teeth were selected on thebasis of 1 of the following: suspected or known to have pulp pathosis(history of trauma, caries, extensive restorations, pain, or discomfort);previously received or currently undergoing endodontic treatment; orprovisionally diagnosed as having a healthy pulp (having no cariesand either minimally restored or with no restorations).

Six of the participants had no restorations in the teeth tested, and in2 of the subjects all of the tested teeth were restored to a certain extent.The teeth morphotypes can be found in Table 1. The state of the pulpsand the presence/absence of restorations are summarized in Table 2.

The order of pulp testing in the first and third visits was LDF, CO2,EF, Ice, EPT. During the second visit, the order of testing was reversed.

Thermal Pulp TestsThese were applied to the gingival third of the buccal surface of

dried and isolated teeth for up to 10 seconds per tooth, with a 2-minute interval between different tests. Alternatively, when any part ofthe tooth had thin enamel or where there was exposed dentin or metalrestorations, these surfaces were used as the test surface if deemedlikely to give better thermal conduction. Subjects were instructed toindicate when they felt the cold sensation, which was then recordedas ‘‘a response’’. Alternatively, ‘‘no response’’ was recorded if the toothfailed to respond twice consecutively in the 1 session. The followingwere used as thermal pulp tests: CO2, carbon dioxide crystals froma pressurized tank compacted in a customized plastic adapter, Odontot-est (Fricar A.G., Zurich, Switzerland) carrier, producing a ‘‘pencil’’ ofdry ice approximately 3.5 mm in diameter; refrigerant spray (EF),Roeko Endo Frost consisting of a butane, propane, and isobutane pres-surized gas mixture in an aluminum can, sprayed onto a no. 2 sizecotton pellet until crystals formed; Ice, empty local anesthetic carpules(Xylocaine; Astra Zeneca, Ryde, Australia) filled with frozen water stan-dardized to 3.5-mm diameter.

TABLE 2. The Restorative and Pulp Status of the Tested Teeth

Restorative statusTeeth with clinically

normal pulps

No restorations 82Minimally restored 10Moderately restored 7Extensively restored 13Total 112

RCT, teeth had root canal treatment commenced but not completed; RCF, teeth had existing root canal fil

1620 Chen and Abbott

For EPT (Electric Pulp Tester; Analytic Technology, Redmond,WA), in accordance with the manufacturer’s instructions, toothpastewas applied to the probe tip to assist electrical conductance. The rateof voltage increase was set between 3 and 4. The probe was placedon the incisal or occlusal third of the tooth where possible. Interprox-imal isolation and insulation with a plastic strip were done wherecurrent conduction to an adjacent tooth through metal was possible.‘‘No response’’ was recorded when no pulp response was noted afterthe digital display reached its maximum level of 80 twice consecutivelyin the 1 session.

In LDF, pulp blood flow (Flux) was measured by a laser Dopplerflowmeter (MoorLAB/FloLAB; Moor Instruments Ltd, Axminster, UK)calibrated against a brownian motion medium and zeroed againsta static reflector. The laser source was 780 nm, with 0.5-mm fiber sepa-ration by using the MoorLAB P13 and P5a probes, with 3.1 kHz as theprimary bandwidth for filter set to 0.1-second time output constant.Customized polyvinylsiloxane (AFFINIS soft/fast putty, AFFINIS regularbody; Colt�ene/Whaledent, Cuyahoga Falls, OH) splints, with holesdrilled for metal sleeves to hold the probe tip in place 2–3mm abovethe buccal gingival margin, were fabricated for each participant, anda 90-second recording time for each tooth was used where possible.Patients were rested in a supine position for 3–5minutes before startingthe LDF test. Signals measured from each tooth were recorded by usingcustomized PC software (LABSOFT; Moor Instruments Ltd) andanalyzed to give the raw data consisting of mean flux value for eachrecording.

To assess whether LDF produced a response, the tested tooth waspaired with its contralateral tooth (except for 1 subject for whom nocontralateral tooth was available) for flux comparison. This was basedon the observation that a tooth with no pulp blood flow has approxi-mately 40% less LDF output signal than a clinically normal pulp (23,25). This can be expressed such that no response was recorded fora tooth if the suspected diseased pulp flux/known healthy pulp fluxratio was #0.6.

The raw data were processed by the software programs MoorSoftfor Windows (Moor Instruments Ltd), Microsoft Excel, Microsoft Word(Microsoft Corporation, Redmond, WA), and SPSS 15.0 (SPSS Inc, Chi-cago, IL).

For comparison of accuracy, reliability, and repeatability, theaverage responses of the patients conducted during 3 visits were sub-jected to interclass correlation (ICC), repeated measures analysis ofvariance (RMANOVA), and pairwise comparisons, with P < .05 as thethreshold of statistical significance.

In this analysis, accuracy was defined as a measure of test effi-cacy explored by RMANOVA and pairwise comparisons examiningbetween-test accuracy differences. Reliability was defined as a measureof test precision, ie, measurement of test efficacy/accuracy during the3 visits. This was explored by RMANOVA and pairwise comparisonsexamining between-test and within-test accuracy differences acrossvisits. Repeatability was defined as a measure of whether a particulartest will yield the same result, irrespective of its accuracy, during the3-visit time frame. ICC was used to explore the repeatability of the

Diseased pulp/RCT/RCF Total

1 830 102 96 199 121

ling.

JOE — Volume 37, Number 12, December 2011

Figure 2. Accuracy of the various pulp tests during 3 visits.

Clinical Research

tests. ICC has a value between 0 and 1, with 1 being most repeatableand 0 being not repeatable.

ResultsThe disease prevalence was approximately 7% (9 of 121 teeth). On

average during the 3 visits, the EPT (0.98) and CO2 (0.97) tests had thehighest accuracy, followed closely by the LDF (0.95) and EF (0.91)tests. The Ice test (0.85) had the lowest overall accuracy (Fig. 1).Summaries of the test accuracies of each visit are provided inFigure 2. RMANOVA showed a significant difference between the accu-racy of the tests (F4,76 = 7.96, mean squared error [MSE] = 0.23, P <.001, h2p = 0.30), ie, some tests were more accurate than others. Pair-wise comparisons indicated that there was no significant difference inaccuracy between the CO2 test, LDF test, and EPT (P > .3). The CO2test was significantly more accurate than the Ice test by 12.3% (P =.005) and EF by 6.3% (P = .022). EPT was significantly more accuratethan Ice by 12.9% (P = .004) and EF by 7.0% (P = .015). The LDF testwas significantly more accurate than Ice by 11.5% (P= .006); althoughthe LDF test was 5.6%more accurate than the EF test, there was no statis-tical significance (P= .063). The EF test was significantly more accuratethan the Ice test by 5.9% (P = .019) but less accurate than CO2 test andEPT, as described above. The Ice test was significantly less accurate thanall other tests (P < .019).

RMANOVA indicated that not all tests were reliable because of thesignificant differences found in between-test accuracy across visits(F2,38 = 8.48, MSE = 0.01, P = .001, h2p = 0.31). Pairwise compari-sons indicated that test accuracies from visit 1 were significantlydifferent to both visit 2 and visit 3 (P= .013 and P= .004, respectively),whereas no statistical difference was found between visit 2 and visit 3 (P= .103). Within-test accuracy across visits showed that results in all 3visits significantly differed in accuracy for the Ice test (F2,38 = 5.67, MSE= 0.01, P = .007, h2p = 0.23), with a maximum difference of 10%across visits. Other tests did not have any significant differences in accu-racy across visits (P> .054); therefore, the Ice test can be considered asbeing relatively unreliable. Pairwise comparisons indicated that theaccuracy for the Ice test from visit 1 was statistically different to visit2 and visit 3 (P= .042 and P= .009, respectively), whereas no statisticaldifference was found between visit 2 and visit 3 (P = .32). For repeat-ability with ICC, the order of the tests from the most to the least repeat-able was Ice, LDF, EF, EPT, and CO2 (Table 3).

DiscussionWhen comparing the current study with other reports, the accu-

racy of pulp tests needs to be interpreted cautiously because otherstudies have reported higher disease prevalence. Petersson et al (26)

Figure 1. Overall average accuracy of the various pulp tests.

JOE — Volume 37, Number 12, December 2011

reported 39% and Evans et al (24) reported 44%, whereas the currentstudy had only 7%. The cohort of subjects and teeth in the study by Fusset al (6) had the same disease prevalence as the current study at 7%, andtheir CO2 test correctly identified teeth with healthy pulps approximately97% of the time. However, it should be noted that no diseased teeth wereincluded in their study. In the current study, the CO2 test correctly iden-tified teeth with no pulp disease 97%of the time, and EPT correctly iden-tified teeth with no pulp disease 98% of the time. In the study by Fusset al (6), EPT identified healthy pulps approximately 90% of the time.This lower figure can be attributed to the inclusion of immature teethin which EPT results have been found to be inconsistent (3, 8). Incontrast to the current study and other studies (2, 23, 27), all 10root-treated teeth tested in the study by Fuss et al (6) did not have falseresponses to EPT.

The EF test identified teeth with healthy or no pulps 91% of the timein this study, which was less than DDM in the study by Fuss et al (6), inwhich healthy pulps were correctly identified 99% of the time. This canbe attributed to the higher temperature of the EF test compared withDDM (28). In vitro studies (6, 12) have reported a faster rate oftemperature change produced by refrigerant sprays than the rate withCO2, but without superior accuracy, and the current study hadsimilar findings with regards to accuracy. Perhaps the warmer oralenvironment might play a more significant effect on test outcomethan was originally suggested by Miller et al (13).

Interestingly, the greatest difference was found with the Ice test,where it correctly identified teeth with healthy or no pulps 85% of

TABLE 3. Test Repeatability ICC

Test type s2p s2

w ICC SE (ICC)

CO2 0.002171 0.00285 0.43233 0.26833Ice 0.02533 0.0121 0.67668 0.22983EF 0.01429 0.01036 0.57952 0.25908EPT 0.002445 0.003189 0.43401 0.26846LDF 0.005086 0.002697 0.65352 0.23847

s2p, between-person variance component; s2w, within-person variance component; SE, standard

error.

Accuracy, Reliability, and Repeatability of Pulp Tests 1621

Clinical Research

the time. Although this was the lowest of the 5 tests in the current study, itis markedly different than the 35% reported by Fuss et al (6). Thisdifference might be due to interindividual patient sensibility differencesas well as the difficulty of controlling the melting ice, which producescold water that flows to other sites (eg, adjacent gingivae, adjacentteeth) and can potentially provide false responses.

The particular LDF testing method used in this study allowed thecorrect identification of all teeth without pulp blood flow on 36 occa-sions. This was achieved by avoiding testing protocols that place limi-tations on LDF readings. This included not using the measurement offlux values from a large pool of teeth as an absolute indicator forquantitative blood flow. This is due to the significant interindividualdifference in flux values (29) as well as the inevitable signal contam-ination from non-pulp sources (30). Also, root-filled teeth were notused as negative controls for flux value. Optical property changes oftooth structure in teeth that have undergone root canal treatment havebeen implicated (25, 31) as being responsible for possible anomaliesin flux readings.

Despite the high accuracy, LDF as a pulp test has the disadvantagesof being time-consuming and technique-sensitive (32) as well as rela-tively expensive (33). In this study, at least 15 minutes per patient wereneeded for the polyvinylsiloxane splint fabrication. If a laboratory-madeacrylic splint is used, the time is estimated to be approximately an hour,and this also increases the costs. In addition, the patients must berelaxed and sitting very still, because suddenmovements can create arti-facts on the flux recordings. Therefore, LDF might not be practical foreveryday clinical use.

The benchmark used in this study for determining whether teethhad pulp blood flow was that teeth without pulp blood flow had at least40% less flux signal than teeth with intact pulp blood flow (diseasedpulp flux/healthy pulp flux#0.6). This was based largely on empiricalobservations made from 2 previous studies (23, 25) that had only 33subjects and slightly differing testing parameters. Further studiesneed to be done to validate this methodology.

A casual observation of this study noted that several pigmentedteeth in 1 patient without restorations gave lower LDF flux readings,similar to the findings of Heithersay and Hirsch (34) in 1 tooth.However, there were insufficient samples in this study for further anal-ysis of the effects of pigmentation. It has been observed that teeth withmore dentin as a consequence of pulp chamber calcification can havehigher LDF flux signals recorded (23). In this study, there was 1 tooththat had significant pulp chamber calcification; this tooth was diagnosedas having a clinically normal pulp, which only responded to EPT, andradiographically normal periapical tissues. The LDF flux readingswere not higher than any of the other teeth tested in the same partici-pant. This participant also underwent external whitening of that partic-ular tooth during the course of 3 weeks (coinciding with this study). Itwas observed that the flux readings of the whitened tooth decreasedduring the 3 testing sessions as the tooth became more opaque. TheLDF flux reading became progressively lower than the contralateraltooth. The effects of stains, discoloration, and dental whitening proce-dures warrant further investigation in future studies.

Accuracy and ReliabilityLDF was statistically indiscriminate (P > .063) to all other tests in

accuracy except when compared with the Ice test (P = .006). The Icetest was significantly less accurate than all other tests (P < .019). TheCO2 and EPT were significantly more accurate than the EF test (P= .022and P = .015, respectively). Although the accuracy of the tests can beranked according to these observations and average values, the reli-ability and repeatability of the tests should also be taken into consider-ation before assessing pulp test performance.

1622 Chen and Abbott

Reliability analysis by using between-test accuracy across visitsindicated that there was no effect as a result of altering the testing order.Given that the testing order was reversed in the second visit and if itsomehow had an effect on the test results, then the test results for thesecond visit should be statistically different to the first and third visits.Fuss et al (6) reported that order reversal of thermal testing and EPTdid not affect their test results, and Pantera et al (35) reported thatEPT reliability was not affected by the use of DDM during testing.

Although RMANOVA of between-test accuracy across visits (F2,38= 8.48, MSE = 0.01, P = .001, h2p = 0.31) suggested that not all thetests were reliable, it should also be noted that there could have beena learning effect with 1 or all of the test modalities, because the firstvisit had lower accuracies compared with the other 2 visits. The exam-ination of this possible effect on a test-by-test basis was shown byintratest reliability.

Only the Ice test showed significant within-test accuracy differ-ences over visits (F2,38 = 5.67, MSE = 0.01, P = .007, h2p = 0.23),where visit 1 was significantly different to visit 2 (P = .042) and visit3 (P = .009). This indicated that the Ice test was not reliable (that is,not consistent in accuracy over 3 visits). However, a learning effect ofthe operator or the participants could not be excluded. The other testsappeared to be relatively reliable (P > .054).

An attempt was made with this analysis to examine whether thetests would yield the same repeatable results irrespective of the accura-cies (Table 3). On a relative scale, Ice and LDF appeared most repeat-able, whereas EPT and CO2 were less repeatable. The ICC values of allthe tests were between 0.4 and 0.7, indicating that no tests in this studywere absolutely repeatable or unrepeatable.

ConclusionsThe accuracy of EPT, CO2, and LDF was 97.7%, 97.0%, and 96.3%,

respectively, with no significant differences in accuracy. The accuracy ofEF and Ice was 90.7% and 84.8%, respectively. EPT and CO2 were signif-icantly more accurate than EF. LDF was more accurate than EF, but thiswas not statistically significant. Ice was significantly less accurate than allthe other tests. With the exception of Ice, all tests were reliable. Ice andLDF were the most repeatable of the tests, whereas EPT and CO2 wereless repeatable. However, Ice was the least accurate and least reliabletest. Within the limitations of this study, no test was superior in all accu-racy, reliability, and repeatability scores compared with the other testswhen analyzed by using different statistical methods. EF was reliable butnot as accurate as EPT and CO2, and it was less repeatable than LDF andIce. CO2, EPT, and LDF were reliable and the most accurate tests, butCO2 and EPT were less repeatable but less time-consuming than LDF.

AcknowledgmentsThe authors are very grateful to the Australian Dental Research

Foundation for their support in this research through the Colin Cor-mie Scholarship.

The authors deny any conflicts of interest related to this study.

References1. Rowe AH, Pitt Ford TR. The assessment of pulpal vitality. Int Endod J 1990;23:

77–83.2. Peters DD, Baumgartner JC, Lorton L. Adult pulpal diagnosis: i—evaluation of the

positive and negative responses to cold and electrical pulp tests. J Endod 1994;20:506–11.

3. Bender IB. Reversible and irreversible painful pulpitides: diagnosis and treatment.Aust Endod J 2000;26:10–4.

4. Ehrmann EH. Pulp testers and pulp testing with particular reference to the use of dryice. Aust Dent J 1977;22:272–9.

JOE — Volume 37, Number 12, December 2011

Clinical Research

5. Mumford J. Thermal and electrical stimulation of teeth in the diagnosis of pulpal

and periapical disease. Proc R Soc Med 1967;60:197–200.6. Fuss Z, Trowbridge H, Bender IB, Rickoff B, Sorin S. Assessment of reliability of

electrical and thermal pulp testing agents. J Endod 1986;12:301–5.7. Bhaskar SN, Rappaport HM. Dental vitality tests and pulp status. J Am Dent Assoc

1973;86:409–11.8. Fulling HJ, Andreasen JO. Influence of maturation status and tooth type of perma-

nent teeth upon electrometric and thermal pulp testing. Scand J Dent Res 1976;84:286–90.

9. Myers JW. Demonstration of a possible source of error with an electric pulp tester.J Endod 1998;24:199–200.

10. Pileggi R, Dumsha TC, Myslinksi NR. The reliability of electric pulp test after concus-sion injury. Endod Dent Traumatol 1996;12:16–9.

11. Davies AL, Rawlinson A. A comparison between two electric vitality testers and ethylchloridewith special reference to a newly available device. Int Endod J 1988;21:320–6.

12. Jones VR, Rivera EM, Walton RE. Comparison of carbon dioxide versus refrigerantspray to determine pulpal responsiveness. J Endod 2002;28:531–3.

13. Miller SO, Johnson JD, Allemang JD, Strother JM. Cold testing through full-coveragerestorations. J Endod 2004;30:695–700.

14. Fuss Z, Trowbridge H, Baker J, Rickoff B, Bender IB. Effects of thermal vitality testson human dental pulp. J Endod 1988;14:482–5.

15. Augsburger RA, Peters DD. In vitro effects of ice, skin refrigerant, and CO2 snow onintrapulpal temperature. J Endod 1981;7:110–6.

16. Peters DD, Mader CL, Donnelly JC. Evaluation of the effects of carbon dioxide usedas a pulpal test: 3—in vivo effect on human enamel. J Endod 1986;12:13–20.

17. Ingram TA, Peters DD. Evaluation of the effects of carbon dioxide used as a pulpal test:part 2—in vivo effect on canine enamel and pulpal tissues. J Endod 1983;9:296–303.

18. Peters DD, Lorton L, Mader CL, Augsburger RA, Ingram TA. Evaluation of the effectsof carbon dioxide used as a pulpal test: 1—in vitro effect on human enamel.J Endod 1983;9:219–27.

19. Emshoff R, Moschen I, Strobl H. Use of laser Doppler flowmetry to predict vitality ofluxated or avulsed permanent teeth. Oral Surg Oral Med Oral Pathol Oral RadiolEndod 2004;98:750–5.

20. Strobl H, Gojer G, Norer B, Emshoff R. Assessing revascularization of avulsed permanentmaxillary incisors by laser Doppler flowmetry. J Am Dent Assoc 2003;134:1597–603.

21. Emshoff R, Emshoff I, Moschen I, Strobl H. Diagnostic characteristics of pulpalblood flow levels associated with adverse outcomes of luxated permanent maxillaryincisors. Dent Traumatol 2004;20:270–5.

JOE — Volume 37, Number 12, December 2011

22. Strobl H, Haas M, Norer B, Gerhard S, Emshoff R. Evaluation of pulpal blood flowafter tooth splinting of luxated permanent maxillary incisors. Dent Traumatol 2004;20:36–41.

23. Ing�olfsson AE, Tronstad L, Hersh EV, Riva CE. Efficacy of laser Doppler flowmetry indetermining pulp vitality of human teeth. Endod Dent Traumatol 1994;10:83–7.

24. Evans D, Reid J, Strang R, Stirrups D. A comparison of laser Doppler flowmetry withother methods of assessing the vitality of traumatised anterior teeth. Endod DentTraumatol 1999;15:284–90.

25. Roebuck EM, Evans DJ, Stirrups D, Strang R. The effect of wavelength, bandwidth,and probe design and position on assessing the vitality of anterior teeth with laserDoppler flowmetry. Int J Paediatr Dent 2000;10:213–20.

26. Petersson K, Soderstrom C, Kiani-Anaraki M, Levy G. Evaluation of the ability ofthermal and electrical tests to register pulp vitality. Endod Dent Traumatol 1999;15:127–31.

27. Bender IB, Landau MA, Fonsecca S, Trowbridge HO. The optimum placement-site ofthe electrode in electric pulp testing of the 12 anterior teeth. J Am Dent Assoc 1989;118:305–10.

28. Herrero de Morais CA, Lima WM, Guerisoli DMZ. Evaluation of the temperature ofdifferent refrigerant sprays used as a pulpal test. Aust Endod J 2008;34:86–8.

29. Norer B, Kranewitter R, Emshoff R. Pulpal blood-flow characteristics of maxillarytooth morphotypes as assessed with laser Doppler flowmetry. Oral Surg Oral MedOral Pathol Oral Radiol Endod 1999;87:88–92.

30. Polat S, Er K, Akpinar KE, Polat NT. The sources of laser Doppler blood-flow signalsrecorded from vital and root canal treated teeth. Arch Oral Biol 2004;49:53–7.

31. Soo-ampon S, Vongsavan N, Soo-ampon M, Chuckpaiwong S, Matthews B. The sour-ces of laser Doppler blood-flow signals recorded from human teeth. Arch Oral Biol2003;48:353–60.

32. Evans D, Reid J, Strang R, Stirrups D. A comparison of laser Doppler flowmetry withother methods of assessing the vitality of traumatised anterior teeth. Endod DentTraumatol 1999;15:284–90.

33. Emshoff R, Emshoff I, Moschen I, Strobl H. Laser Doppler flow measurementsof pulpal blood flow and severity of dental injury. Int Endod J 2004;37:463–7.

34. Heithersay GS, Hirsch RS. Tooth discoloration and resolution following a luxationinjury: significance of blood pigment in dentin to laser Doppler flowmetry readings.Quintessence Int 1993;24:669–76.

35. Pantera EA Jr, Anderson RW, Pantera CT. Reliability of electric pulp testing afterpulpal testing with dichlorodifluoromethane. J Endod 1993;19:312–4.

Accuracy, Reliability, and Repeatability of Pulp Tests 1623