Time-course of the effect of potassium oxalate in the treatment of hypersensitive dentine in man.

OBJECTIVE: Investigate the cause of hypersensitive dentine (HD) by recording the time course of changes in dentine sensitivity, sensory threshold to electrical stimulation (ET) and pulpal blood flow (PBF) following tubular occlusion using in vitro and clinical experiments. DESIGN: Nineteen teeth with HD and 13 with normal dentine from 8 participants were evaluated, and the intensity of any pain produced by various stimuli was recorded at different times after oxalate treatment. The participants used a visual-analogue scale (VAS) to indicate the intensity of any pain. The ET and PBF were recorded at the same times. RESULTS: Preliminary in vitro experiments showed that oxalate treatment had no effect on the method used to record PBF, and blocked the treated tubules immediately after application. Considering teeth with HD, a decrease in the median VAS evoked by all forms of stimulation was observed at all post-treatment times, except immediately after treatment (p < 0.05), while the treatment produced no significant effect in teeth with normal dentine. No significant changes in ET or PBF was observed in any of the groups. CONCLUSIONS: The effect of oxalate in relieving the symptoms of HD is not only due to a reduction in the intensity of stimulation of sensory receptors sensitive to fluid flow in the dentinal tubules, but also to a reduce in the sensitivity of the receptors that respond to dentine stimulation. There was no evidence that acute pulpitis or central sensitization to pain, which would be associated with changes in PBF or ET, contributes to HD.

Electroosmosis in human dentine in vitro.

OBJECTIVE: To determine the rate of fluid flow through human dentine due to electroosmosis during iontophoresis of either 2 % lignocaine with epinephrine, Ringer's solution, epinephrine, or distilled water. DESIGN: Experiments were carried out on 24 intact extracted human premolars. Dentine was exposed at the tip of the buccal cusp. The cavity was filled with one of the test solutions and the pulp cavity, with Ringer's solution at a pressure of 11 mm Hg. Fluid flow through the dentine was measured using a capillary connected to the pulp cavity. Current was passed between a stainless-steel electrode in the cavity and one in the pulp cavity. The results were analysed using repeated measures, three-way ANOVA, with Bonferroni pairwise comparisons where this showed a significant effect. RESULTS: The current passed produced a significant flow of fluid through the dentine but neither the composition of the test solution nor etching had a significant effect on the flow. During iontophoresis of 2 % lignocaine with epinephrine for example, currents of 0.2, 0.4, and 0.6 mA applied with the cavity electrode as anode produced inward flow rates of 2.25 ± 0.87, 5.00 ± 1.62, 8.60 ± 1.97 (mean ± s.d.) nL/s/mm² respectively, and applying the currents in the opposite direction caused outward flows of 0.76 ± 0.72, 1.00 ± 1.01, 1.12 ± 1.18 nL/s/mm² respectively. CONCLUSIONS: It is concluded that electroosmosis can be produced in human dentine, it can enhance the effect of iontophoresis in transporting charged molecules through dentine, particularly large molecules, and it could also enable uncharged molecules to be carried through dentine into the pulp.

Pulpal blood flow recorded from exposed dentine with a laser Doppler flow meter using red or infrared light.

OBJECTIVE: To determine the percentage of the blood flow signal that is derived from dental pulp when recording from exposed dentine in a human premolar. DESIGN: Recordings were made from 7 healthy teeth in 5 subjects (aged 22-33 yr.) with a laser Doppler flow meter (Periflux 4001) using either a red (635 nm) or an infrared (780 nm) laser. After exposing dentine above the buccal pulpal horn (cavity diam. 1.6 mm, depth 3 mm) and isolating the crown with opaque rubber dam, blood flow was recorded alternately with infrared or red light from the exposed dentine under four conditions: before and after injecting local anaesthetic (3% Mepivacaine without vasoconstrictor) (LA) over the apex of the root of the tooth; after exposing the pulp by cutting a buccal, class V cavity in the tooth; and after sectioning the coronal pulp transversely through the exposure. RESULTS: There was no significant change in mean blood flow recorded with either light source when the tooth was anaesthetized or when the pulp was exposed. After the pulp had been sectioned, the blood flow recorded with infrared light fell by 67.8% and with red light, by 68.4%. The difference between these effects was not significant. CONCLUSIONS: When recording blood flow from exposed coronal dentine with either infrared or red light in a tooth isolated with opaque rubber dam, about 68% to the signal was contributed by the pulp. The signal:noise ratio was better with infrared than red light, and when recording from dentine than enamel.

Observations on fluid flow from exposed dentine in primary teeth: An in vitro study.

OBJECTIVE: To investigate fluid flow through dentine in primary teeth in vitro using the replica technique, and to compare the results with those obtained from permanent dentine. DESIGN: The experiments were carried out on 22 extracted, mandibular, primary, incisor teeth. The incisal edge was removed to 1mm below the dentino-enamel junction and half the exposed surface etched with phosphoric acid. The exposed dentine was blotted dry and the pressure in the pulp cavity held at 0, 15, 30 or 45cm H2O above atmospheric for 30s. Fluid that accumulated on the dentine surface was recorded with impression material and a replica made with epoxy resin which was examined in a scanning electron microscope. RESULTS: Structures resembling fluid droplets were present in the replicas of unetched dentine in all 22 teeth, and at all the pulpal pressures tested. The droplets formed at 45cm H2O were significantly larger (median diam., 5.14mm; interquartile range, 3.26mm; Friedman repeated measures analysis of variance on ranks (RMAVR) and Tukey test) than those formed at other pressures. There was no evidence of droplets in the replicas of etched dentine with any of the pulpal pressures. CONCLUSIONS: These results demonstrate that fluid will tend to flow from dentine in deciduous teeth when it is exposed. They are similar to those obtained in a previous study in this laboratory on permanent teeth. The fact that fluid droplets were absent from etched dentine suggests that, after being blotted, the etched dentine matrix absorbed fluid that tended to flow out through the dentinal tubules.

Effect of full crown preparation on pulpal blood flow in man.

OBJECTIVE: To determine if full crown preparation causes an increase in pulpal blood flow (PBF), indicating inflammation, in human subjects. DESIGN: The experiments were carried out on 35 intact, mandibular posterior teeth in 13 subjects: 32 were abutments for 16 fixed bridges that replaced first molars; the other 3 were first premolars adjacent to abutment teeth that served as un-operated controls. Crown preparations were made using an air-rotor with water-spray under regional block anaesthesia (4% articaine with epinephrine 1:100,000). PBF was recorded with a laser Doppler flow meter (LDF) before and after administering the anaesthetic, with the LDF probe on the buccal enamel. PBF was then recorded from the abutment teeth with the probe on buccal dentine after preparing the buccal surfaces of both teeth, after completing the crown preparations, and after 1 and 7days. PBF was also recorded from the buccal enamel of the control teeth on each occasion. RESULTS: The mean±S.D. PBF values before and after anaesthesia were 2.63±2.13 and 2.42±2.38P.U. respectively, which were not significantly different (Paired t-test). The mean values for the abutment teeth after buccal preparation, after complete crown preparation, and after 1 and 7days were 5.20±2.49, 4.53±2.52, 4.92±2.98 and 5.48±2.65P.U. respectively. The 4 values for each tooth were not significantly different (two-way RM ANOVA). In the control group, the values under all six conditions were not significantly different. CONCLUSIONS: Regional block anaesthesia produced no change in PBF, nor did full-crown preparation, neither immediately after the procedure nor 1 and 7days later.

Effect of cavity depth on dentine sensitivity in man.

OBJECTIVE: To determine if dentine at the enamel-dentine junction (EDJ) in man is more sensitive to hydrostatic pressure stimuli then deeper dentine. DESIGN: Cavities (1mm diam.) were cut at the tips of the buccal and lingual cusps of 8 premolars in 3 subjects (ages: 22-25 years). Both cavities were initially deepened to expose the EDJ then one (the test cavity) was deepened in steps of 0.5mm to a maximum of 2.0 mm below the EDJ. The cavities were tested at each stage, before and after etching, with 5s, hydrostatic pressure stimuli between 400 mm above, and 400 mm below atmospheric. The intensity of any pain produced was recorded on a VAS scale and electrodes were placed in both cavities in an attempt to monitor any action potentials evoked in intradental nerves. RESULTS: In all the teeth, the intensity of the pain produced by a stimulus tended to increase as the cavity was deepened, as did the number of action potentials recorded (in 6 of the 8 teeth). The responses were greater from etched than unetched dentine, and negative pressures evoked greater responses than the corresponding positive pressures. CONCLUSION: There was no evidence that dentine close to the EDJ was more sensitive to hydrostatic pressure stimuli than deeper dentine. It may however be more sensitive to mechanical stimuli as it is more compliant.

Effects of bleaching on laser Doppler blood-flow signals recorded from human teeth in vitro.

OBJECTIVE: To determine if bleaching a tooth affects significantly laser Doppler recordings of pulpal blood flow that are made through the treated enamel and dentine. DESIGN: Recordings of blood flow (flux) were made with a laser-Doppler flow meter from the crowns of 14, recently extracted, human premolars while diluted blood was pumped at three different rates through a cannula that was inserted into the pulp cavity with its tip in the coronal pulp chamber. The recordings of flux were made before and after bleaching the crown with 38% hydrogen peroxide for 30min. The colour changes produced by bleaching were also measured with a digital spectrophotometer in 15 teeth. The effects of bleaching on both blood flow recordings and tooth colour were studied in 11 of the teeth. RESULTS: Both before and after bleaching, there was a linear relationship between blood perfusion rate and flux. After bleaching, the flux readings increased by an average of 46%, and were significantly increased in every tooth (P<0.001, Wilcoxon Signed Rank Test). There was no significant correlation between the changes in flux and the changes in tooth colour. CONCLUSION: A laser Doppler flow meter cannot be used to determine the effect of bleaching on pulpal blood flow in intact teeth.

Effects of the iontophoresis of lignocaine with epinephrine into exposed dentine on the sensitivity of the dentine in man.

OBJECTIVE: To determine the effects of the iontophoretic application of lignocaine and epinephrine to exposed dentine on the sensitivity of the dentine in human subjects. DESIGN: The experiments were carried out on 13 healthy premolars (13 subjects) that were scheduled for extraction. Dentine was exposed at the tip of the buccal cusp by cutting a cavity which was etched with 35% phosphoric acid. The sensitivity of the dentine was tested with probing and air blast stimuli. The subject indicated the intensity of any pain produced with a score of 0-100. In 7 teeth, the cavity was filled with a solution containing 20% (w/v) lignocaine HCl and 0.1% (w/v) epinephrine HCl, and an iontophoretic current of 120 µA was passed for 90s. The sensitivity of the dentine was tested before and immediately after the treatment and then at 10 min. intervals for 40 min. Pulpal blood flow was recorded at each stage. Control experiments were carried out on 6 teeth using a solution containing only the epinephrine. RESULTS: The lignocaine plus epinephrine solution completely blocked the pain produced by both forms of stimulus immediately, and this continued for at least 40 min. It also produced an immediate fall in pulpal blood flow that also lasted for at least 40 min. The epinephrine solution had the same effect on pulpal blood flow but no effect on dentine sensitivity. CONCLUSIONS: The topical application of 20% lignocaine and 0.1% epinephrine, with an iontophoretic current of 120µA for 90s, will anaesthetize exposed, normal, dentine.

The iontophoresis of lignocaine with epinephrine into carious dentine for pain control during cavity preparation in human molars.

OBJECTIVE: To determine the effectiveness of the iontophoretic delivery of lignocaine with epinephrine through carious dentine for pain control during cavity preparation. DESIGN: The experiments were carried out on 56 carious molars that required class I restorations in 42 subjects (aged 15-20 years). The overhanging enamel and soft caries were removed then the sensitivity of the exposed dentine was tested with drilling, probing and air blast stimuli. The subject indicated the intensity of any pain produced by marking a visual analogue scale (VAS). The cavity was then filled with 20% w/v lidocaine with 0.1% w/v epinephrine and a 200 µA iontophoretic current applied for 2 min after which the sensitivity of the dentine was re-tested. If the dentine was not anaesthetized, the treatment and testing were repeated up to 6 times. RESULTS: The total duration (min) of iontophoresis required to anaesthetize the dentine was: 2 in 7 teeth, 4 in 17 teeth, 6 in 14 teeth, 8 in 4 teeth, and 10 in 7 teeth. The remaining 7 teeth were not anaesthetized even after 14 min of iontophoresis. CONCLUSIONS: The iontophoretic delivery of lignocaine with epinephrine anaesthetized dentine for cavity preparation in 49 of 56 (87.5%) of carious molars.

Effect of the topical application of 50% lignocaine hydrochloride on the sensitivity of dentine in man.

OBJECTIVE: To determine the effect of topical applications of 50% (w/v) lignocaine HCl on the sensitivity of human dentine. DESIGN: The experiments were carried out on 12 premolars scheduled for extraction as part of orthodontic treatment in nine subjects (ages: 16-29 years). Dentine was exposed by cutting a cavity at the tip of the buccal cusp of each tooth, and etched with 35% phosphoric acid. The sensitivity of the exposed dentine to probing and air-blast stimuli was assessed before and after applying either 50% (w/v) lignocaine HCl solution or distilled water to the exposed dentine for 10min. Changes in the sensitivity of the dentine were monitored for up to 160min. The subject indicated the intensity of any pain produced by marking a 100mm visual analogue scale (VAS). RESULTS: Before treatment, both forms of stimulus evoked pain in all the teeth. The median VAS score with probing was 40mm and, with air-blast stimulation, 30mm. 50% lignocaine HCl produced a progressive fall in these scores and after 30min there was no response to either probing or air-blast stimulation. The responses started to return 30-160min after the lignocaine had been washed off. Water had no effect. CONCLUSIONS: Lignocaine will diffuse into exposed dentine and block the pain evoked by probing and air-blast stimuli provided that a sufficiently steep diffusion gradient is created. A topical application of a 50% (w/v) solution of lignocaine HCl for 10min will anaesthetise dentine within 30min.

A comparison between red and infrared light for recording pulpal blood flow from human anterior teeth with a laser Doppler flow meter.

OBJECTIVE: To compare red (635 nm) and infrared (780 nm) light for recording pulpal blood flow from human anterior teeth with a laser Doppler flow meter. DESIGN: Recordings were made from 7 healthy teeth in 5 subjects (aged 22-55 years) using a laser Doppler flow meter (Periflux 4001) equipped with both red and infrared lasers. Average blood flow signals were obtained with both light sources alternately from each tooth under five conditions: intact tooth without opaque rubber dam, intact tooth with dam, after injecting local anaesthetic (3% Mepivacaine) (LA) over the apex of the tooth and cavity preparation to almost expose the pulp, after removal and replacement of the pulp, and with the root canal empty. RESULTS: With infrared light, because of technical limitations, data were obtained for the first three conditions only. The dam significantly decreased the mean blood flow by 82%. Injecting LA and cavity preparation had no significant effect. With red light, dam produced a decrease of 56%, and the resulting signal was reduced by 33% after LA and cavity preparation. The remaining signal fell by 46% after pulp removal and replacement. This contribution of the pulp is similar to that recorded previously with infrared light. There was no significant further change when the pulp was finally removed. CONCLUSIONS: The importance of using opaque rubber dam is confirmed. With dam, there is no advantage to using red rather than infrared light, and in each case the pulp contributes less than 50% to the blood flow signal.

Pulpal blood flow recorded from human premolar teeth with a laser Doppler flow meter using either red or infrared light.

OBJECTIVE: To compare red (635 nm) and infrared (780 nm) light for recording pulpal blood flow from human premolar teeth. DESIGN: Recordings were made from 11 healthy teeth in 9 subjects (aged 16-30 years) using a laser Doppler flow meter (Periflux 4001) equipped with both red and infrared lasers. Average blood flow signals were obtained with both light sources alternately from each tooth under five conditions: intact tooth without opaque rubber dam, intact tooth with dam, after injecting local anaesthetic (3% Mepivacaine) (LA) over the apex of the tooth and cavity preparation to almost expose the pulp, after removal and replacement of the pulp, and with the root canal empty. RESULTS: With infrared light, the dam significantly decreased the mean blood flow by 80%. Injecting LA and cavity preparation had no significant effect. Removal and replacement of the pulp reduced the mean blood flow by 58%. There was no further change when the pulp was removed. With red light, the dam reduced the signal from intact teeth by 60%. Injecting LA and cavity preparation had no significant effect. The signal fell by 67% after pulp removal and replacement and did not change significantly when the pulp was removed. CONCLUSIONS: Opaque rubber dam minimises the contribution of non-pulpal tissues to the laser Doppler signal recorded from premolars. Using dam, the pulp contributed about 60% to the blood flow signal with both red and infrared light. The difference between them in this respect was not significant.

The effect of applying potassium chloride solutions at atmospheric pressure on the sensitivity of dentine in man.

OBJECTIVE: To determine the effects of applying KCl solutions to exposed dentine at atmospheric pressure on pain evoked by probing and air blast stimuli in human subjects. DESIGN: The experiments were carried out on 24 premolars in 16 subjects (aged 16-30 years). A cavity (diameter 3mm, depth 3mm) was cut at the tip of the buccal cusp and etched with 35% orthophosphoric acid. The cavity was filled with 500, 250 mmol/l KCl or 500 mmol/l NaCl for 10 min, after which it was rinsed with normal saline. Each solution was tested in 8 teeth. Air blast and probing stimuli were applied to the exposed dentine prior to dentine treatment and at 0, 2, 5, 10, and 20 min after treatment. After each stimulus the subject indicated the intensity of the pain evoked on a visual analogue scale (VAS). RESULTS: In teeth treated with 250 or 500 mmol/l KCl, the mean VAS response to air blast stimuli was significantly decreased at 5 and 10 min after treatment. The mean VAS response to probing was significantly decreased 10 min after treatment with 500 mmol/l KCl. Otherwise there were no significant changes. CONCLUSION: Topical application to exposed dentine of solutions containing a high concentration of potassium ions at atmospheric pressure produce a temporary reduction in the sensitivity of dentine to air blast and probing stimuli.

Atrial fibrillation associated with a thyroid stimulating hormone-secreting adenoma of the pituitary gland leading to a presentation of acute cardiac decompensation: a case report.

INTRODUCTION: Hyperthyroidism is a well established cause of atrial fibrillation (AF). Thyroid Stimulating Hormone-secreting pituitary tumours are rare causes of pituitary hyperthyroidism. Whilst pituitary causes of hyperthyroidism are much less common than primary thyroid pathology, establishing a clear aetiology is critical in minimising complications and providing appropriate treatment. Measuring Thyroid Stimulating Hormone (TSH) alone to screen for hyperthyroidism may be insufficient to appropriately evaluate the thyroid status in such cases. CASE PRESENTATION: A 63-year-old Caucasian man, previously fit and well, presented with a five-day history of shortness of breath associated with wheeze and dry cough. He denied symptoms of hyperthyroidism and his family, social and past history were unremarkable. Initial investigation was in keeping with a diagnosis of atrial fibrillation (AF) with fast ventricular response leading to cardiac decompensation.TSH 6.2 (Normal Range = 0.40 - 4.00 mU/L), Free T3 of 12.5 (4.00 - 6.8 pmol/L) and Free T4 51(10-30 pmol/L). Heterophilic antibodies were ruled out. Testosterone was elevated at 43.10 (Normal range: 10.00 - 31.00 nmol/L) with an elevated FSH, 18.1 (1.0-7.0 U/L) and elevated LH, 12.4 (1.0-8.0 U/L). Growth Hormone, IGF-1 and prolactin were normal. MRI showed a 2.4 cm pituitary macroadenoma. Visual field tests showed a right inferotemporal defect.While awaiting neurosurgical removal of the tumour, the patient was commenced on antithyroid medication (carbimazole) and maintained on this until successful trans-sphenoidal excision of the macroadenoma had been performed. AF persisted post-operatively, but was electrically cardioverted subsequently and he remains in sinus rhythm at twelve months follow-up off all treatment. CONCLUSION: This case reiterates the need to evaluate thyroid function in all patients presenting with atrial fibrillation. TSH-secreting pituitary adenomas must be considered when evaluating the cause of hyperthyroidism. Early diagnosis and treatment of such adenomas is critical in reducing neurological and endocrine complications.

Effects of potassium ions on dentine sensitivity in man.

OBJECTIVE: To determine the effect of applying 500 mmol/l KCl at a pressure of 150 mm Hg above atmospheric to exposed dentine on pain sensation evoked by probing and air blast stimuli in human subjects. DESIGN: The experiments were carried out on 14 pairs of premolars in 14 human subjects (aged 17-30 years). Dentine was exposed at the tip of the buccal cusp, etched with acid and covered with saline, then 500 mmol/l KCl in Ringer's was applied at a pressure of 150 mm Hg for 4 min. Mechanical probing and air-drying stimuli were performed before and 2, 10, 20, 30 min after applying the KCl solution. The subject indicated the intensity of any pain produced on a visual analogue scale (VAS). Pulpal blood flow was recorded with a laser Doppler flow meter. Exactly the same procedure was carried out on the contralateral tooth except that 500 mmol/l NaCl in Ringer's was used in place of the KCl solution. RESULTS: The pain responses to mechanical probing and air blast stimuli were significantly reduced during the first 10 min after applying the KCl but not the NaCl solution. Pulpal blood flow did not change significantly after either treatment. CONCLUSION: Potassium ions, when applied to exposed dentine at a pressure of 150 mm Hg above atmospheric, produced temporary block of impulse conduction in sensory nerve endings in the dentine or pulp.

Pain and the rate of dentinal fluid flow produced by hydrostatic pressure stimulation of exposed dentine in man.

OBJECTIVE: To determine the relationship between pain intensity and the rate of fluid flow through dentine in human subjects. DESIGN: The experiments were carried out on 16 premolars in 13 human subjects (aged 15-25 years). Dentine was exposed at the tip of the buccal cusp, etched with acid and covered with saline. A series of 5 s hydrostatic pressure stimuli between 400 mmHg above and 400 mmHg below atmospheric were applied to the dentine, in steps of 50 mmHg. The subject indicated the intensity of any pain produced on a visual analogue scale (VAS). The fluid flow through dentine during application of the same stimuli was measured in vitro within 3 h after tooth extraction. RESULTS: The median pain threshold with negative (subatmospheric) stimuli was -125 mmHg and, with positive pressure stimuli, 200 mmHg, which corresponded to dentinal fluid flow rates of 3.29 nL/(s mm(2)) exposed dentine and 5.75 nL/(s mm(2)), respectively. Both the median pressure and the mean rate of flow at threshold with negative pressures were significantly lower than with positive pressures. The curves relating VAS score to stimulus intensity were similar with both negative and positive pressures. CONCLUSION: The sensory transduction mechanism for pain in human teeth is more sensitive to outward than inward flow through dentinal tubules. The difference in sensitivity was however much less than that of the hydrodynamic receptors in the cat, which respond very much more strongly to negative than positive pressure stimuli.

The relationship between the discharge of intradental nerves and the rate of fluid flow through dentine in the cat.

OBJECTIVE: To investigate further the relationship between dentinal tubular flow and the discharge evoked in intradental nerves. DESIGN: In anaesthetised cats, recordings were made of fluid flow through dentine during the application of hydrostatic pressure stimuli of 5 s duration in the range +500 to -500 mm Hg to exposed dentine and of the nerve impulses evoked by these stimuli. Single unit recordings were obtained from filaments dissected from the inferior alveolar nerve and multi-unit recordings, from the exposed dentine. RESULTS: Of 20 single units tested, 10 (conduction velocities: 2.4-36.2 m s(-1)) responded to negative pressures and four of these, also to positive pressures. None responded to only positive pressures. The pressure thresholds of the units (single and multi-unit preparations) ranged from -100 to -500 and +100 to +500. In terms of flow (measured 1 s after the start of a stimulus) the thresholds ranged from 0.4 to 2.2 nl s(-1) mm(-2) exposed dentine with outward flow, and 0.4-2.1 nl s(-1) mm(-2) with inward flow. The outward flow per tubule at the threshold of the most sensitive units was estimated to be 21 fl s(-1) and the corresponding mean velocity of the contents of the dentinal tubules at their pulpal ends, 27 microm s(-1). Although the thresholds to outward and inward flow were similar, with outward flow the mean discharge rate increased with stimulus intensity; whereas with inward flow few impulses were evoked and the number was little affected by the stimulus intensity. CONCLUSION: The transduction mechanism that generates impulses in hydrodynamic intradental afferents is much more responsive to outward than inward flow through the dentinal tubules, although the thresholds in both directions are similar.

Sensory transduction mechanisms responsible for pain caused by cold stimulation of dentine in man.

OBJECTIVE: To determine the effects on the sensitivity of exposed dentine to cold that are produced when dentine is etched to remove the smear layer and when the tubules are blocked again with calcium oxalate. Separate in vitro observations were made on the effects of these procedures on fluid flow through the dentine. DESIGN: The experiments were carried out on 24 premolars in 17 subjects. Dentine was exposed at the tip of the buccal cusp and cold stimuli were applied by placing the tip of an ice stick on the cavity floor for 5s under the following conditions: before etching the dentine, after etching, and after oxalate treatment. The subject indicated the intensity of any pain produced on a visual analogue scale (VAS). Fluid flow through the dentine was recorded under similar conditions in eight of the teeth in vitro. RESULTS: The mean VAS score produced by the ice before etching was 21.3+/-19.5mm (S.D.). This increased significantly to 85.4+/-15.6mm after etching (P<0.01). After oxalate treatment, it decreased significantly to 8.5+/-13.3mm. The corresponding mean rates of fluid flow through dentine were 2.15+/-1.02, 1.55+/-0.84, and 2.29+/-1.28nL/smm(2) exposed dentine, respectively. The mean after etching was significantly less than the other two values (P<0.05). CONCLUSION: If the pain was due to hydrodynamic receptors, their sensitivity to dentinal fluid flow changed when the tubules were opened or closed. Alternatively the pain was produced by receptors sensitive to some other change produced by the cold stimuli, such as specific cold receptors.