Clinical Research

Evaluation of Dental Pulp Sensibility Tests in a Clinical Setting

James J. Jespersen, DDS,* John Hellstein, DDS, MS,† Anne Williamson, DDS, MS,*
William T. Johnson, DDS, MS,* and Fang Qian, PhD‡

Abstract
Introduction: The goal of this project was to evaluate
the performance of dental pulp sensibility testing with
Endo Ice (1,1,1,2-tetrafluoroethane) and an electric
D iagnosis is often as much an art as it is a science. Clinicians must combine the results
of their clinical exam, radiographic findings, reported dental history, and clinical
experience to arrive at an accurate diagnosis (1). Although pulp sensibility testing is an
pulp tester (EPT) and to determine the effect of several essential part of endodontic diagnosis and has been extensively studied, no strong cor-
variables on the reliability of these tests. Methods: Data relation between results and certain key factors has been established (2–7). This
were collected from 656 patients seen in the University research project was designed to gather data in a clinical setting with a large sample
of Iowa College of Dentistry Endodontic graduate clinic. size to facilitate a better understanding of the accuracy of sensibility tests and factors
The results of pulpal sensibility tests, along with the that may affect their outcome.
tooth number, age, sex, number of restored surfaces, Much of the uncertainty involved with pulpal diagnosis is the fact that the dental
presence or absence of clinical or radiographic caries, pulp is encased in hard tissue and not available for direct examination. Surrogate tests
and reported recent use of analgesic medications, must therefore be used, with an ultimate goal of extrapolating the histologic status of the
were recorded. The presence of vital tissue within the pulp tissue from the results. Although commonly referred to as vitality tests, the most
pulp chamber was used to verify the diagnosis. Results: common clinical procedures used to determine pulpal status are more accurately
The Endo Ice results showed accuracy, 0.904; sensitivity, named pulpal sensibility tests. This is because they are not a direct measure of pulpal
0.916; specificity, 0.896; positive predictive value, vitality, but instead they rely on a subjective response by the patient to an external stim-
0.862; and negative predictive value, 0.937. The EPT re- ulus to the nervous system (8, 9). The most commonly used pulpal sensibility tests are
sults showed accuracy, 0.75; sensitivity, 0.84; speci- cold and electrical pulp testers (EPT). Our goal was to evaluate the performance of
ficity, 0.74; positive predictive value, 0.58; and Endo Ice (EI) (1,1,1,2-tetrafluoroethane; Hygenic Endo-Ice Green; Colt
ene Whaledent,
negative predictive value, 0.90. Patients aged 21–50 Cuyahoga Falls, OH) and EPT (SybronEndo Vitality Scanner; SybronEndo, Glendora,
years exhibited a more accurate response to cold testing CA) in direct patient care and examine the effects of tooth type, age, sex, restorative sta-
(P = .0043). Vital teeth with caries responded more tus, caries, and recent analgesic use on diagnostic accuracy.
accurately to cold testing (P = .0077). There was no sta-
tistically significant difference noted with any other var- Materials and Methods
iable examined. Conclusion: Pulpal sensibility testing Study design was approved by the Institutional Review Board at the University of
with Endo Ice and EPT are accurate and reliable methods Iowa. Standardization of testing procedures, data gathering protocols, and clinical
of determining pulpal vitality. Patients aged 21–50 ex- criteria was accomplished for all research participants before the start of the study.
hibited a more accurate response to cold. Sex, tooth Data were collected from patients seen by graduate students in the Endodontic Depart-
type, number of restored surfaces, presence of caries, ment at the University of Iowa College of Dentistry. A total of 656 patients were included,
and recent analgesic use did not significantly alter the with an age range of 6–85 years (Table 1). Only those tests that were deemed necessary
results of pulpal sensibility testing in this study. (J Endod for an accurate diagnosis were performed before treatment. All 656 patients were tested
2014;40:351–354) with EI. Seventy-seven patients required the additional use of EPT for an accurate diag-
nosis. Because of the subjective nature of pulp sensibility tests, only 1 tooth per patient
Key Words was included in the data set to maximize the statistical power of the results.
Electric pulp tester, Endo Ice, pulp testing, sensibility
testing, vitality testing Exclusion Criteria
 The inability to perform or adequately respond to pulp testing (this might include a
patient with intellectual disabilities, a patient who presents to the clinic previously
From the *Department of Endodontics, †Department of Oral anesthetized from another provider, or a patient seen under general anesthesia)
Pathology, Radiology and Medicine, and ‡Biostatistics Unit,
University of Iowa College of Dentistry, Iowa City, Iowa.  Trauma within the last 6 months (9, 10)
Address requests for reprints to Dr Anne Williamson, Dental  Previous root canal treatment or initiation of treatment
Science Building, Department of Endodontics, 801 Newton  Inability to determine the presence or absence of bleeding pulp in the pulp chamber,
Road, Iowa City, IA 52242. E-mail address: such as in cases of severe pulp canal obliteration
[email protected]
0099-2399/$ - see front matter All teeth were adequately dried and isolated before testing. Adjacent teeth were
Copyright ª 2014 American Association of Endodontists. tested as controls before the tooth being treated to observe a baseline normal
http://dx.doi.org/10.1016/j.joen.2013.11.009
response. Cold testing with EI was accomplished by using a large cotton pellet
on the buccal surface of the tooth for 15 seconds, or until patient indicated a
response (11). EPT testing was done with dentifrice used as a conductive medium
(12). The probe was placed on intact tooth structure, and a response before an 80

JOE — Volume 40, Number 3, March 2014 Pulp Sensibility Tests 351
Clinical Research
TABLE 1. Demographics There was no significant correlation noted with gender, tooth type
Age group (y) or location, number of restored surfaces, or recent analgesic use. No
6–20 172 significant difference was noted for age or the presence of caries with
21–35 185 EPT. Molars were somewhat more likely to exhibit simultaneous false
36–50 107 positives to both EI and EPT (P = .0993).
51–65 126
66–85 66
Gender
Male 299 Discussion
Female 357 The overall accuracy and statistical data in this study gathered with
Tooth location both cold and EPT are listed in Table 3, along with comparison of the
Maxillary anterior 89
Maxillary premolar 79 results of 2 comparable studies. EI results showed that it was accurate
Maxillary molar 221 and reliable approximately 90% of the time it was used. EPT was accu-
Mandibular anterior 32 rate 75% of the time and was used in generally more difficult diagnostic
Mandibular premolar 39 situations. Comparison of the results between studies should focus on
Mandibular molar 196
the values of sensitivity and specificity. These terms are prevalence-
independent descriptors, and their results are an intrinsic property of
reading was recorded as positive. All teeth included in the study un- the test itself, whereas accuracy and predictive values can vary with dis-
derwent endodontic treatment, and the presence or absence of ease prevalence.
bleeding pulp in the pulp chamber on access was used as a true pos- The presence or absence of tissue in the pulp chamber was used as
itive or true negative. Bleeding tissue in both the chamber and canal a measure of vitality (3, 4, 6, 7). This was taken as a valid measure of
space(s) was required for a designation of vital. After treatment, the pulpal status because it can be expected that the majority of the insults to
patient’s age, sex, tooth, presence of caries noted either clinically or the pulp occur from the clinical crown. If coronal necrosis occurs, it
radiographically, number of restored surfaces, reported analgesic will proceed in an apical direction, and the absence of pulp in the
use within the last 24 hours, pulp test results, and presence of tissue chamber is a sign of at least partial necrosis. Necrosis of any part of
in the pulp chamber were recorded. the pulp tissue is an irreversible condition that will require root
The accuracy, sensitivity, specificity, positive predictive value canal treatment to avoid apical periodontitis if the tooth is to be
(PPV), and negative predictive value (NPV) for both cold and EPT retained, and so it is a valid measure of pulpal status (13).
were calculated and are summarized in Table 2. Accuracy was examined Refrigerant spray is the most common method of pulp testing used
for correlation to the variables listed above. Age correlation was accom- clinically because it is convenient, easy to use, and reliable (14).
plished by using 15-year age ranges. Tooth type comparison was done Because of its popularity, variables with significant correlation to testing
by comparing different groupings of teeth including tooth type (molar, results would be highly applicable to clinical practice. Our study deter-
premolar, etc), maxillary versus mandibular, anterior versus posterior, mined age and the presence of caries in vital teeth significantly affected
and multirooted versus single-rooted. All data analysis was accom- the accuracy of testing with EI. Table 4 shows the results of cold with
plished by using SAS for Windows (v9.3; SAS Institute Inc, Cary, NC). regard to age and that patients aged 21–50 have a statistically signifi-
Within the context of this study, accuracy is the proportion of sen- cantly more accurate response to cold when compared with those in
sibility test results that agreed with a direct visualization of the pulp other age categories (P = .0043). Although other studies have exam-
chamber. Sensitivity describes the proportion of vital teeth that are accu- ined age as a potential variable, this was the first to have the sample
rately identified as vital by the sensibility test. For instance, if 10 teeth size large enough to determine a difference (4). An explanation for
known to be vital are tested and 9 respond to testing, the sensitivity of reduced accuracy in young patients may be that pulpal innervation is
that test would be 0.9. Conversely, specificity describes the proportion not complete until late in root development (15, 16). Younger
of necrotic teeth that are correctly identified as necrotic. The PPV de- patients may also be more anxious and less reliable because of the
scribes the proportion of teeth identified as vital by sensibility testing subjective nature of the tests. Decreasing pulpal innervation and
that are actually vital when accessed. For example, if 10 teeth test positive increasing pulpal mineralization have also been shown with age (17,
and 8 teeth are found to actually be vital when accessed, the PPV of that 18). This general decrease in innervation, along with additional
test would be 0.8. The NPV is similar to PPV; however, NPV describes the secondary or tertiary dentin, may help explain why sensibility testing
proportion of teeth identified as nonvital that are actually nonvital. In all is less accurate as age increases older than 50.
categories, a result of 1.0 shows 100% accuracy, 0.5 shows the test is The presence of caries in vital teeth was significantly associated
correct 50% of the time, and 0.00 indicates the test is never correct. with more accurate testing with EI (P = .0077). A pulpal response to
caries begins as early as white spot lesions limited to enamel, with
increasing inflammation as the carious lesion progresses toward the
Results pulp (19, 20). It is likely that the increased responsiveness to cold
The presence of caries in vital teeth resulted in a more accurate seen in this study is a consequence of inflammation from an innate
response to cold testing (P = .0077). immune response resulting in decreased action potential thresholds
Patients aged 21–50 responded more accurately to cold testing (21). Although there was a marginal correlation (P = .0778) of
(P = .0043). increased accuracy to cold when all teeth were examined, this was

TABLE 2. Performance of Cold and EPT

Cold EPT

Accuracy Sensitivity Specificity PPV NPV Accuracy Sensitivity Specificity PPV NPV
0.904 0.916 0.896 0.862 0.937 0.753 0.840 0.742 0.583 0.902

352 Jespersen et al. JOE — Volume 40, Number 3, March 2014

 Clinical Research
TABLE 3. Comparison with Recent Studies
Cold EPT

Study Accuracy Sensitivity Specificity PPV NPV Accuracy Sensitivity Specificity PPV NPV
Jespersen (n = 656/77) 0.904 0.916 0.896 0.862 0.937 0.753 0.840 0.742 0.503 0.902
Weislider (n = 150) 0.83* 0.76 0.92 0.93 0.74 0.81* 0.92 0.75 0.83 0.87
Petersson (n = 59) 0.86 0.83 0.93 0.89 0.90 0.81 0.72 0.93 0.88 0.84
*Personal communication, October 2012, Daniel J. Caplan, DDS, PhD.

exclusively attributable to vital teeth because there was no comparable a definite tendency toward false-positive responses with EPT. Data
correlation with necrotic teeth. It is important to consider that although showing that vital teeth will accurately respond to at least one test, while
caries does appear to make vital teeth respond more accurately to cold, at the same time indicating the most common error encountered is false
pulpal status cannot always be known with certainty at the time of positives, does not lend itself to a clear recommendation or interpreta-
testing. With this in mind, a positive response to cold should most likely tion of results. Instead, it emphasizes the need for the clinician to use all
not be interpreted differently in the presence of caries; however, no the data available to them, along with their clinical expertise and expe-
response to EI with a carious tooth may make a diagnosis of pulpal ne- rience, to render a diagnosis.
crosis more certain. Restorative status, sex, reported recent use of analgesic, and tooth
In this study, EI was accomplished for every patient, whereas EPT type or location resulted in no significant differences with regard to
was done at the discretion of the provider treating the case. Generally, pulpal sensibility testing. These findings are consistent with previous
this means EPT was used in challenging diagnostic situations where studies (4, 26–28). Tooth type was analyzed in a variety of
there were discrepancies between the patient’s history, radiographic, combinations, and although no correlation to tooth type was noted
and clinical findings. Lower overall values calculated for EPT compared with regard to individual tests, molars were found to be marginally
with EI is a reflection of the more difficult cases in which it was used. In more likely to exhibit simultaneous false positives to both EI and EPT
clinical practice, EPT may be used more frequently in cases that are not (P = .0993). This finding is comparable to Peters et al (29), who found
adequately diagnosed with EI as the sole sensibility testing agent. There- previously treated molars with an unobturated canal to be more likely to
fore, data regarding its performance under these conditions may be have false-positive response to EPT. Multiple roots and an increased
highly applicable to its most common clinical usage. volume of pulp tissue may make molars more likely to be in a state
EPT use in this study showed a large discrepancy between PPV and of partial necrosis at the time of testing, resulting in a greater chance
NPV. The high NPV calculated for EPT means that a negative reading can of false-positive results.
confidently be interpreted as pulpal necrosis, whereas the low PPV in- There are many potential variables that were not addressed by this
dicates a predilection for false-positive readings. Possible causes of false study. These could include the presence or absence of preoperative
positives to EPT include adjacent restorations, vital nerve fibers after pain, correlation with periapical test results, duration of response to
coronal necrosis, or stimulation through adjacent periodontal attach- thermal testing, and correlation with radiographic findings, among
ment (22–24). It has been suggested that EPT may actually result in others.
fewer false-negative responses when compared with cold in older pa- More research is needed with key variables seen on a daily basis in
tients, because sclerotic dentin may impede the flow of dentinal fluid the practice of endodontics and their effect on the validity and accuracy
within the tubules (25). However, results from this study do not support of dental sensibility tests.
this conclusion. Our data suggest that a negative reading on EPT is an The ultimate goal of diagnosis is to extrapolate the histologic
excellent predictor of necrosis, whereas a positive reading is a relatively status of the pulp from clinically available data. However, even tests
poor predictor of vitality. that directly measure pulpal blood flow have yet to reach a level of
Weisleder et al (3) showed that the most accurate diagnosis was sophistication that would allow them to definitively relay this informa-
achieved when a combination of test results was used. This was not tion. Although visual examination of the chamber shows the presence
necessarily the case in our study. Table 5 shows the results of the 77 of tissue, the pulp can undergo asymptomatic irreversible inflamma-
teeth that were tested with both EI and EPT. In contrast to the study tion while still containing bleeding tissue within the chamber (30).
by Weisleder et al, we found only 67.5% agreement between the tests The inclusion of histologic data as well as more detailed information
with correct results. One hundred percent of vital teeth tested with EI regarding patient responses in large-scale sensibility testing studies
and EPT responded to at least one of the sensibility tests, which is illus- would provide the most accurate and clinically applicable data
trated in the data presented in Table 5 showing that no tooth responded possible. With this in mind, the figures reported here are an oversim-
with false-negative responses to both cold and EPT. However, there was plification of the accuracy of dental sensibility tests. More research

TABLE 5. Comparison of Cold and EPT Data for Teeth Tested with Both Cold
TABLE 4. Performance of EI versus Age and EPT
Correct Incorrect False False
Age group (y) (n = 593) (n = 63) P value Total positive negative
Test (n = 77) (test) (test)
6–20 150 (87.2%) 22 (12.8%) .0043*
21–35 174 (94.1%) 11 (5.9%) Cold and EPT both correct 52 NA NA
36–50 100 (93.5%) 7 (6.5%) Cold correct EPT incorrect 12 8 (EPT) 4 (EPT)
51–65 113 (89.7%) 13 (10.3%) EPT correct Cold incorrect 6 3 (Cold) 3 (Cold)
66–85 56 (84.9%) 10 (15.1%) Both incorrect 7 7 (Both) 0 (Both)
*Mantel-Haenszel c2 test. NA, not applicable.

JOE — Volume 40, Number 3, March 2014 Pulp Sensibility Tests 353
Clinical Research
concerning the histologic status of the pulp and its correlation with 11. Jones DM. Effect of the type carrier used on the results of dichlorodifluoromethane
pulpal sensibility testing is required to increase the accuracy of application to teeth. J Endod 1999;25:692–4.
12. Mickel AK, Lindquist KA, Chogle S, et al. Electric pulp tester conductance through
dental pulp sensibility tests. various interface media. J Endod 2006;32:1178–80.
13. Langeland K. Tissue response to dental caries. Dent Traumatol 1987;3:149–71.
Conclusions 14. Hargreaves K, Cohen S, eds. Cohen’s Pathways of the Pulp, 10th ed. St Louis, MO:
Mosby; 2010:2–39.
EI accuracy was 0.904. EPT accuracy was 0.75. Patients aged 21– 15. Johnsen D, Harshbarger J, Rymer H. Quantitative assessment of neural development
50 exhibited a more accurate response to cold testing. Vital teeth re- in human premolars. Anat Rec 1983;205:421–9.
sponded more accurately to cold. No significant difference was seen 16. Johnsen D. Innervation of teeth: qualitative, quantitative, and developmental assess-
with any other variable examined. More research is required to accu- ment. J Dent Res 1985;64:555–63.
rately correlate the histologic status of the pulp with the response to 17. Swift M, Byers M. Effect of ageing on responses of nerve fibres to pulpal inflamma-
tion in rat molars analysed by quantitative immunocytochemistry. Arch Oral Biol
dental sensibility testing. 1992;37:901–12.
18. Bernick S, Nedelman C. Effect of aging on the human pulp. J Endod 1975;1:
Acknowledgments 88–94.
19. Brannstrom M, Lind PO. Pulpal response to early dental caries. J Dent Res 1965;44:
The authors thank the AAE foundation for their support. 1045–50.
The authors deny any conflicts of interest related to this study. 20. Trowbridge HO. Pathogenesis of pulpitis resulting from dental caries. J Endod 1981;
7:52–60.
21. Hahn C, Liewehr FR. Innate immune responses of the dental pulp to caries. J Endod
References 2007;33:643–51.
1. Bender I. Pulpal pain diagnosis: a review. J Endod 2000;26:175–9. 22. Myers JW. Demonstration of a possible source of error with an electric pulp tester.
2. Mejare I, Axelsson S, Davidson T, et al. Diagnosis of the condition of the dental pulp: J Endod 1998;24:199–201.
a systematic review. Int Endod J 2012;45:597–613. 23. Torneck CD. Changes in the fine structure of the human dental pulp subsequent to
3. Weisleder R, Yamauchi S, Caplan DJ, et al. The validity of pulp testing: a clinical carious exposure. J Oral Pathol Med 1977;6:82–95.
study. J Am Dent Assoc 2009;140:1013–7. 24. Nahri M, Virtanen A, Kuhta J, Huopaniemi T. Electrical stimulation of teeth with a
4. Peterson K, S€oderstr€om C, Kiani-Anaraki M, Levy G. Evaluation of the ability of ther- pulp tester in the cat. Eur J Oral Sci 1979;87:32–8.
mal and electrical tests to register pulp vitality. Dent Traumatol 1999;15:127–31. 25. Ehrmann EH. Pulp testers and pulp testing with particular reference to the use of dry
5. Chen E, Abbott PV. Evaluation of accuracy, reliability, and repeatability of five dental ice. Aust Dent J 1977;22:272–9.
pulp tests. J Endod 2011;37:1619–23. 26. Miller SO, Johnson JD, Allemang JD, Strother JM. Cold testing through full-coverage
6. Villa-Chavez CE, Pati~no-Marın N, Loyola-Rodrıguez JP, et al. Predictive values of restorations. J Endod 2004;30:695–700.
thermal and electrical dental pulp tests: a clinical study. J Endod 2013;39:965–9. 27. Kardelis AC, Meinberg TA, Sulte HR, et al. Effect of narcotic pain reliever on pulp
7. Dastmalchi N, Jafarzadeh H, Moradi S. Comparison of the efficacy of a custom-made tests in women. J Endod 2002;28:537–9.
pulse oximeter probe with digital electric pulp tester, cold spray, and rubber cup for 28. Fuss Z, Trowbridge H, Bender IB, et al. Assessment of reliability of electrical and
assessing pulp vitality. J Endod 2012;38:1182–6. thermal pulp testing agents. J Endod 1986;12:301–5.
8. Chen E, Abbott PV. Dental pulp testing: a review. Int J Dent 2009;12:365785. 29. Peters DD, Baumgartner JC, Lorton L. Adult pulpal diagnosis: I—evaluation of the
9. Levin LG. Pulp and periradicular testing. J Endod 2013;39:S13–9. positive and negative responses to cold and electrical pulp tests. J Endod 1994;20:
10. DiAngelis AJ, Andreasen JO, Ebeleseder KA, et al. International association of dental 506–11.
traumatology guidelines for the management of traumatic dental injuries: 1—frac- 30. Lin L, Langeland K. Light and electron microscopic study of teeth with carious pulp
tures and luxations of permanent teeth. Dent Traumatol 2012;28:2–12. exposures. Oral Surg Oral Med Oral Pathol 1981;51:292–316.

354 Jespersen et al. JOE — Volume 40, Number 3, March 2014