Premedication with the 20-40 mg oral piroxicam has shown efficacy in preventing spontaneous POP particularly with symptomatic irreversible pulpitis [13,14,15,16]; the efficacy of 20 mg sublingual piroxicam FDT on spontaneous compared to stimulated postoperative pain as well as RAI and FUI was, yet, to be investigated in cases of non-vital pulp which can be more likely to experience postoperative pain rise.

Asymptomatic teeth with non-vital pulp were included for being more susceptible to pain rise after treatment [6, 7, 19]. Also, preoperative pain strongly predicts postoperative pain [3, 4, 20]; including asymptomatic teeth could, thus, avoid its confounding effect allowing more effective differentiation of the effect of piroxicam administration on different postoperative pain types. Mandibular molars were associated with more postoperative pain [3, 6]. This study, thus, represented a ‘worst case scenario’ for postoperative pain experience. A standardized protocol was implemented to minimize the effect of possible intraoperative confounders [21, 22]. Moderate certainty exists that single-visit treatment with AP can provide better radiographic outcome than multiple-visit treatment [23].

Various tools have been used for pain assessment of which NRS shows higher compliance rates, responsiveness, with high reliability and validity [24]. Patients were trained to use NRS to reduce measurement bias [21]. A 7-day trial duration, compared to a maximum of 3 days in previous studies [13,14,15,16], can be sufficient to allow the manifestation of the preemptive effects of piroxicam, the longest-acting NSAID used in endodontics [10] if they exist. Most studies assess spontaneous rather than stimulated postoperative pain [6, 10, 19]. Stimulated pain can be assessed on chewing, biting and/or percussion; a bite test can better simulate patient’s pain on chewing than percussion test [6, 19].

This study was a double-blind, randomized placebo-controlled, clinical trial with an allocation ratio of 1:1 where block randomization and blinding should reduce allocation bias risk, performance and ascertainment bias respectively [21]. In the present study, baseline characteristics were balanced between the two groups.

Compared to previous studies including asymptomatic teeth with necrotic pulp, POP characteristics (Intensity and/or incidence) in their control groups were either comparable to [25], higher than [17, 26] or lower than [27, 28] the findings of the control group in this study across corresponding timepoints. Discrepancies could be rendered to variations in population’s criteria, e.g. sex, tooth type, periapical radiolucency, in treatment protocols implemented e.g. number of visits, instrumentation kinematics, occlusal reduction, and/or pain scales [17, 26,27,28].

Piroxicam showed less severity and incidence of POP within the first 24 h, POPB for the first week, and POPer at day 7, and less RAI incidence. So the null hypothesis was rejected. This was in partial agreement with previous studies [13,14,15,16] where piroxicam showed efficacy for 24 h [14, 15], 48 h [13], or 72 h [16]. Differences in efficacy duration could be rendered to variations in patients’ criteria, in treatment protocols e.g. dose [13] and administration time, use of placebo [13, 16], endodontic procedures [14, 15], in pain type assessed, and/or methodological differences e.g. follow-up duration and randomization [14]. Most previous studies included patients with symptomatic irreversible pulpitis [13, 15, 16], and posterior teeth [13,14,15].

Apparent differences could be detected for the trend over time for postoperative pain (POP and POPB) between piroxicam and placebo groups where a significant rise occurred with placebo compared to piroxicam. This was in disagreement to the results of previous relevant studies assessing piroxicam efficacy [13,14,15]; all included symptomatic teeth so only a decline in pain incidence and/or intensity over time from the preoperative pain levels was reported. The findings of this study, particularly of the control group, are in agreement with previous studies with asymptomatic teeth, where a rise in pain levels can occur in the first 24 h [2, 6, 17, 25,26,27,28]. The extended piroxicam efficacy up to 7 days with POPB compared to 24 h with POP could be attributed to the higher severity of POPB and for a longer duration than POP (Table 3, Fig. 2A and B) as well as possible preemptive analgesic effects [18].

Piroxicam, an enolic-acid derivative, is a non-selective NSAID with the main mechanism of action being the inhibition of the cyclooxygenase enzymes, resulting in reduced prostaglandin synthesis [11]. Piroxicam may, also, inhibit activation and aggregation of neutrophils, hence, implying additional mechanisms of action including decreasing proinflammatory cytokine levels [11]. Piroxicam inhibits thromboxane synthesis in platelets, thus, inhibits the secondary phase of platelet aggregation; given the role of platelets in the inflammatory process, this action may contribute to the efficacy of piroxicam [29]. Piroxicam has the longest half-life (50 h) of all NSAIDs due to a low systemic clearance rate with 99% protein binding; time to peak plasma drug concentration after a single dose is 3-5 h [11]; taken together, this could explain its ability to inhibit pain rise and/or flare-up which reached its maximum at 6 h in this study and maintain such pain inhibition throughout the study duration. Pretreatment analgesia before root canal treatment may decrease the establishment of peripheral and central sensitization which has the potential to reduce postoperative pain and rescue analgesic intake [18].

Few studies have assessed stimulated postendodontic over time [6, 20]. Stimulated pain is usually more severe and lasts longer than spontaneous pain [6, 19, 30]; this was supported by the findings of this study where POPB showed higher severity and longer duration than POP (Table 3, Fig. 2B). Henry et al. (2001), however, showed similar spontaneous and stimulated pain values [18]. In this study, POPB levels were within the mild range and incidence range between 28 and 49% throughout 7 days in the control group. Lower stimulated pain levels, however, were reported by Jang et al. (2021), probably due to their pain assessment after the last visit of 2–3 visits and performing occlusal reduction [6]. Henry et al. (2001) reported higher incidence range (41% to 91%) in patients with ‘symptomatic’ teeth and assessing ‘percussion’ pain [19, 20].

Within the limits of this study, premedication seemed to be the most prominent factor associated with POP and POPB overwhelming the effect of other factors (Table 4). Food intake tended to affect POP at 24 h and POPB at 24 h and 48 h at 10% level of significance where Fasting patients had less pain intensity than those who Ate (Table 4). This could be due to the presence of food whichincreases the mean time to reach maximum plasma concentration compared to the fasting state [31, 32].

The findings of the present study could be limited by the relatively small sample size, which could affect the precision of the estimates. Only mandibular molars were included which could affect the generalizability to other teeth types, however, results of this study could simulate the worst case scenario. Performing randomization, patient- and operator-blinding together with absence of loss to follow up could be considered strengths in this study improving its internal validity. The choice of patient-relevant and reported outcome measures, particulay those related to pain on function, is integral to enhance generalizability. Including asymptomatic teeth at baseline helped investigate different types of postoperative pain without the confounding effect of preoperative pain. Future clinical trials with larger sample sizes, including patients with different maxillary and mandibular teeth types, and comparing piroxicam with other analgesic are recommended.