Diagnostic Accuracy of Cold, Heat, and Electric Pulp Tests

This is a prospective, in vivo, observational diagnostic accuracy study conducted at the Endodontics Clinic of Istanbul University Faculty of Dentistry. The protocol was approved by the Ethical Committee of Istanbul University Faculty of Dentistry (Ref. 2024/48 Rev-2) and registered in the Turkish Medicines and Medical Devices Agency (Protocol No. 2023/38). The study was performed in accordance with the Declaration of Helsinki, and written informed consent was obtained prior to study procedures. Before study initiation, the research team was trained and calibrated for the testing protocol and data collection.

Eligible adult participants presenting with at least one tooth indicated for root canal treatment were recruited during routine clinical care. One tooth per participant was included. Eligibility criteria and exclusions were defined a priori and are described in the Eligibility module.

Index Tests (Pulp Sensibility Tests) All index tests were performed by a single operator who was blinded to the participants' clinical signs/symptoms, dental history, and radiographic findings. Participants were instructed to raise their hand immediately when they perceived pain, discomfort, or tingling.

To establish an individual baseline response, testing was first performed on an asymptomatic, functional contralateral tooth without prior endodontic treatment; if unavailable, an adjacent tooth meeting these criteria was used. The study (index) tooth was tested last. Teeth were isolated with cotton rolls and dried with an air syringe before each test. Tests were applied to standardized anatomical sites (incisors/canines: middle third of buccal surface; premolars: occlusal third; molars: mesiobuccal cusp tip area or the closest available site). Responses were recorded dichotomously as "response present" or "response absent" (response time and intensity were not measured). The three tests were performed in the following order, with a 5-minute interval between tests to minimize carryover effects:

Cold test: Refrigerant spray containing propane-butane (Endo-Frost; Roeko, Coltene Whaledent, Germany) was applied to a size 2 cotton pellet. Once frosting was visible, the pellet was placed on the standardized site for up to 15 seconds or until the participant responded.

Heat test (frictional heat): A sterile rubber cup (NAIS, Sofia, Bulgaria) mounted on a low-speed contra-angle handpiece at a constant speed setting was applied with light, consistent contact without water cooling. The stimulus was stopped when the participant responded or after a maximum of 7 seconds.

Electric pulp test (EPT): An apex locator with integrated pulp tester (Ai-Pex; Woodpecker, Guilin, P.R. China) was used with toothpaste as a conducting medium on a 2-mm electrode tip. The current was increased at the lowest ramp setting, and the numeric value at first perception was recorded. The contralateral tooth served as the control. If the response for the index tooth occurred at a value equal to or lower than the control, the result was recorded as "response present"; if a higher value was required or no response occurred up to the maximum value (80), the result was recorded as "response absent."

Reference Standard (Pulpal Bleeding) Immediately after completion of the index tests, local anesthesia (Ultracain D-S Forte; Sanofi, Paris, France) was administered and rubber dam isolation was applied. A standard endodontic access cavity was prepared by a second researcher who was blinded to the pulp test results. The presence or absence of pulpal bleeding in the pulp chamber was assessed by direct visual inspection under 3.5× magnification (dental loupes) and recorded as present/absent. Teeth were classified as vital when bleeding was observed within the pulp chamber. Teeth were classified as non-vital when no bleeding was present in the pulp chamber. Cases with bleeding limited to root canal(s) without pulp chamber bleeding, or bleeding not present in all canals of multi-rooted teeth, were classified as partially necrotic and recorded as non-vital. Teeth with uncertain pulp status were excluded. Following reference standard assessment, root canal treatment was performed as planned.

Statistical Approach For each index test, true-positive, false-positive, true-negative, and false-negative values were determined against the reference standard. Sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy were calculated and reported with 95% confidence intervals (Wilson score method). Agreement between each test and the reference standard was assessed using Cohen's kappa. Differences among the three tests were evaluated using Cochran's Q test, and pairwise comparisons were performed using Bonferroni-adjusted Z tests. Analyses were conducted using IBM SPSS Statistics (Version 23; IBM Corp., Armonk, NY, USA), with statistical significance set at p<0.05.

Inclusion criteria

• Age ≥ 18 years
• At least one tooth indicated for root canal treatment confirmed at the Endodontics initial examination unit
• Able and willing to provide written informed consent

Exclusion criteria

Tooth-related

• Full-coverage crown
• Extensive restoration affecting testing area
• History of dental trauma (for the study tooth)
• Root resorption
• Incomplete root development (open apex)
• Tooth fracture or crack
• Regressed pulp chamber / pulp canal obliteration or calcified root canals Patient-related
• Severe or uncontrolled systemic disease: ASA physical status 3-6