Perioperative loxoprofen sodium is associated with lower pain and anxiety levels in dental implant surgery: a cohort study in China

Abstract

Objectives

To evaluate the effect of a perioperative regimen of 60 mg loxoprofen sodium in Chinese patients who underwent single-tooth implant placement.

Materials and methods

362 patients were allocated to either the loxoprofen group or the control group. Inverse probability of treatment weighting (IPTW) was applied to balance seven baseline covariates. Postoperative infection rate, perceived pain intensity, anxiety levels, and overall patient satisfaction were collected and analyzed.

Results

The loxoprofen group demonstrated significantly lower VAS pain scores at 30 min preoperative (after premedication), on the evening of implant surgery, and at 24 h after implant surgery (p < 0.001). No significant difference in trait anxiety (T-AI) scores was observed (p = 0.426). State anxiety (S-AI) scores were significantly lower in the loxoprofen group at all three time points (p < 0.001). Postoperative infection rates did not differ significantly (p = 0.266). Patient satisfaction was significantly higher in the loxoprofen group (p < 0.001). Subgroup analyses confirmed beneficial treatment effects across age, sex, BMI, ASA status, smoking status, alcohol consumption, and implant sites.

Conclusions

Perioperative administration of loxoprofen sodium was associated with significantly lower pain and state anxiety levels, as well as improved patient satisfaction. The analgesic and anxiolytic effects were most pronounced in the evening of the surgery day. The clinical utility of loxoprofen sodium should be further evaluated in randomized controlled trials.

Clinical relevance

A perioperative regimen of loxoprofen sodium significantly reduces pain and state anxiety, and improves patient satisfaction. It may be an effective analgesic in implant dentistry in China, and multicenter long-term randomized controlled studies are needed.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00784-026-07092-3.

Introduction

Dental implant surgery has evolved into a predictable, high-survival procedure for tooth replacement, with the average success rates exceeding 98% [1]. Yet despite advances in surgical technique and implant biomaterials, perioperative pain remains an unresolved problem. The implant surgical sequence involves reflection of mucoperiosteal flaps, osteotomy preparation with rotary instruments, and fixture placement, all of which release significant amounts of inflammatory mediators that activate peripheral nociceptors and central sensitization pathways [2, 3]. Clinical studies have demonstrated that pain-related inflammatory mediators typically peak and remain intense during the first 24 h postoperatively, with the evening of surgery representing the period of greatest analgesic demand [4, 5]. The somatosensory alterations following implant placement further complicate the perioperative pain experience [6, 7]. Unresolved inflammation and bacterial colonization have been implicated in the development of peri-implant mucositis and subsequent peri-implantitis [8, 9].

Loxoprofen sodium, a prodrug belonging to the nonsteroidal anti-inflammatory drug (NSAID) class, achieves peak plasma concentrations within 30 to 50 min in humans [10, 11]. Its unique prodrug property offers potential gastrointestinal benefits compared to traditional NSAIDs, making it an attractive option for perioperative pain control [12]. In fact, loxoprofen is commonly prescribed for pain management in China [13, 14]. It has been reported that loxoprofen can reduce pain after tooth extraction [15, 16], but no large-scale study has specifically evaluated its efficacy in implant surgery.

Dental anxiety represents a very common and overlooked dimension of perioperative care [17, 18]. Dental anxiety can directly heighten pain perception through central sensitization mechanisms, while uncontrolled pain exacerbates anxiety and creates a detrimental cycle that compromises recovery and patient satisfaction [19–21]. Therefore, effective analgesia may confer anxiolytic benefits beyond simple pain relief [22, 23]. Actually, anti-inflammatory agents might play a role in psychiatric disorders like depression and anxiety [24, 25].

Although NSAIDs like loxoprofen sodium are widely prescribed for dental implant surgery in China, evidence supporting loxoprofen sodium’s use in this specific context remains scarce. Given the distinct pharmacokinetic and prescribing patterns in Asian populations, a rigorous evaluation in a Chinese cohort is essential to inform clinical practice [26, 27]. This observational study employing inverse probability of treatment weighting was designed to investigate whether a standardized perioperative regimen of loxoprofen sodium significantly reduces pain and anxiety levels following single-tooth implant placement, whether this regimen decreases the risk of postoperative infection and ultimately results in enhanced overall patient satisfaction in the Chinese population. We hypothesized that loxoprofen sodium would provide effective symptom management and improve patient satisfaction without compromising surgical safety.

Materials and methods

Study design and patient selection

This was a single-center, retrospective cohort study conducted at a stomatological hospital in China, between April 2024 and October 2025. The study was conducted in accordance with the principles of the Declaration of Helsinki and the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines. Patients who underwent single-tooth implant placement surgeries during the study period were identified from the institutional electronic medical record system. Inclusion criteria were: (1) age 20 to 69 years; (2) American Society of Anesthesiologists (ASA) physical status I or II; (3) had completed medical and pharmacy records; and (4) availability of pain and anxiety assessments during the perioperative period.

Patients who meet any of the following criteria were excluded: (1) known allergy or contraindication to NSAIDs; (2) concurrent use of anticoagulants or antiplatelet agents; (3) history of radiotherapy to the head and neck region; (4) serious systemic diseases (e.g.: diabetes mellitus with HbA1c > 9.0%, uncontrolled hypertension); (5) pregnancy or lactation; (6) psychiatric disorders affecting objective self-evaluation; and (7) use of any analgesic medication other than loxoprofen sodium (e.g.: acetaminophen, ibuprofen) one week before or after the implant surgery. They had no prior implant experience and underwent single-tooth implant placement without bone grafting. A trapezoidal flap design was used for all the patients.

We recruited patients who visited our hospital between April 2024 and October 2025, ensuring sufficient statistical power for the analysis. A total of 407 eligible patients were initially identified, exceeding the required sample size. Among them, 43 patients were using anticoagulant drugs, among whom 36 patients had serious systemic diseases, and two other patients were allergic to NSAIDs. After excluding these 45 patients, 362 patients were finally included in the study. Based on medication administration records, the 362 patients were allocated to two cohorts. Patients in the loxoprofen group (n = 127) received a 60 mg loxoprofen sodium tablet orally 30 min before implant surgery, on the evening of surgery, and 24 h postoperatively respectively. Dose 1 was administered by surgeons 30 min preoperatively (after premedication). Doses 2 and 3 were self-administered by patients at home on the evening of surgery and 24 h postoperatively. To verify adherence, all patients were provided with a medication instruction card to remind them of the time each dose. At the 7-day follow-up visit, research assistants confirmed with patients whether they have taken the medicine. Only patients who took three doses on time were included in the loxoprofen group. Patients in the control group (n = 235) did not receive any prescribed analgesic medication during the perioperative period.

Based on previous literature related to pain, the MCID for the VAS scale was conservatively set at 1.0 point, representing the smallest pain reduction perceived as beneficial by patients [28, 29]. The pooled standard deviation (SD) was estimated as 1.5, derived from historical medical data from our hospital. These data yielded a standardized effect size (Cohen’s d) of 0.67. Using this conservative effect size, a two-sided t-test was used to calculate the required sample size. With a significance level (α) of 0.05 and a power of 0.90, the required sample size was calculated to be 98 patients, with the allocation ratio 1:1. However, the actual allocation ratio was approximately 1:1.85 (loxoprofen: control). Under this allocation ratio between the two groups, the required sample sizes were 69 patients in the control group and 37 in the loxoprofen group, for a total of 106 patients. Our final cohort exceeded the minimum requirement. Figure 1 indicates the flow chart diagram of the study.

Fig. 1.

Flow chart diagram of the study

All surgical procedures were performed by four professional surgeons, and standardized surgical protocols were always followed. Patients gargled with 0.12% chlorhexidine gluconate before implant surgery. Local anesthesia was administered using 4% articaine hydrochloride and epinephrine tartrate injection. After implant surgery, all patients were instructed to use ice packs. Sutures were removed on postoperative day 7.

Outcome measures and data collection

The primary outcome variable was pain intensity, assessed using the Visual Analogue Scale (VAS) [30]. Patients were instructed to mark their pain levels at three time points: (a) 30 min preoperative in the waiting room; (b) the evening of surgery; (c) 24 h postoperative. VAS score 0 means no pain, 1–3 means mild pain, 4–6 means moderate pain, and 7–10 means severe pain.

Secondary outcomes were postoperative infection, anxiety level, and patient satisfaction. Patients were assessed on postoperative day 7. Patients who presented at least one of the following criteria were considered to have postoperative infection: (1) surgical wound dehiscence with purulent exudate; (2) fistula formation with suppuration; (3) abscess formation requiring incision and drainage; (4) persistent redness and swelling that did not respond to conservative management and required antibiotic therapy. The use of antibiotics without accompanying clinical signs of infection was not considered a criterion of postoperative infection. Antibiotics were used only in suspected infection cases and were strictly reserved for the treatment of confirmed postoperative infection.

Patient satisfaction was assessed on the seventh day after implant surgery using a 5-point Likert scale [31, 32]. Patients rated their overall satisfaction based on their overall perioperative experience, including pain management, communication with staff, and overall comfort. The State-Trait Anxiety Inventory (STAI) was used to evaluate anxiety levels, which consists of the State Anxiety scale (S-AI, 20 items) and the Trait Anxiety scale (T-AI, 20 items). Each item is scored from 1 to 4, with total scores for each subscale ranging from 20 to 80. Higher scores indicate greater anxiety [33, 34]. A STAI score of 20–40 means no obvious anxiety, 41–44 means mild anxiety, 45–60 means moderate anxiety, and 61–80 means severe anxiety. It is worth mentioning that T-AI was administered only at baseline (30 min preoperatively) to assess patients’ stable anxiety predisposition. S-AI was administered at the same three time points as VAS to capture perioperative fluctuations of anxiety levels. VAS and S-AI assessments at the “30 minutes preoperative” time point were conducted after the preoperative loxoprofen sodium dose.

Data were extracted from electronic medical records and patient questionnaires by two independent researchers. We selected patient-level indicators as covariates, including age, sex, body mass index (BMI), ASA status, smoking status, alcohol consumption, and implant site. The smoking history refers to whether the patient has ever smoked. Drinking alcohol refers to whether the patient drank alcohol three months before implant surgery.

Statistical analysis

To mitigate confounding of covariates in retrospective observational designs, inverse probability of treatment weighting (IPTW) was applied. The propensity score conditional on observed baseline covariates was estimated using multivariable logistic regression. Standardized mean differences (SMD) less than 0.10 were considered indicative of acceptable balance. Love plots were constructed to visualize the balance of covariates before and after IPTW. Multivariable regression models with IPTW were used to analyze data. For NNT calculations, VAS scores were dichotomized into moderate-to-severe pain (VAS ≥ 4) and mild/no pain (VAS < 4). S-AI scores were dichotomized into moderate-to-severe anxiety (S-AI ≥ 45) and no/mild anxiety (S-AI < 45). IPTW-weighted logistic regression models were fitted for these binary outcomes at each time point. The risk of the control group subtracts the risk of the loxoprofen group was the absolute risk reduction. NNT was derived as the reciprocal of the absolute risk reduction. 95% confidence intervals of NNT were calculated using the non-parametric bootstrap method. Exploratory visualizations of the multidimensional outcome data are provided using principal component analysis (PCA) and a radar chart.

Subgroup analyses were performed to evaluate treatment effects across different populations. Analyzed subgroups included age (< 45 VS ≥ 45 years), BMI (< 22 VS ≥ 22 kg/m²), sex, ASA status, smoking status, alcohol consumption, and implant site. To assess whether the treatment effect of loxoprofen sodium varied according to baseline patient characteristics, we tested interactions between treatment effects and covariates. For VAS pain scores at three time points, S-AI scores at three time points, T-AI score and patient satisfaction, we fitted IPTW-weighted linear regression models which included interaction terms between treatment and covariates. Interaction terms were examined separately for each covariate, and the p-values were used to determine whether each covariate modified the treatment effect. For exploratory analyses like subgroup analyses, interaction tests, NNT calculations, PCA, and radar charts, no multiplicity adjustment was applied to avoid overly conservative type II errors.

In order to assess the robustness of our findings to model specification, we conducted a doubly robust sensitivity analysis. The doubly robust sensitivity analysis combines IPTW with outcome regression, providing consistent treatment effect estimates if either the propensity score model or the outcome regression model is correctly specified. Outcome regression models were fitted with the same covariates as IPTW. All statistical tests were two-tailed, and p < 0.05 was defined as statistically significant. Analyses were performed using R version 4.5.1.

Results

Patient characteristics and propensity score balancing

Table 1 presents baseline characteristics of 362 patients before and after weighting. After IPTW, all covariates in Table 1 achieved excellent balance, with SMD values less than 0.1 for all seven variables. The Love plot (Supplementary Fig. 1) visualizes this substantial improvement in covariate balance.

Table 1.

Baseline characteristics before and after IPTW

Characteristic	Before IPTW			After IPTW
	Loxoprofen group	Control group	SMD	Loxoprofen group	Control group	SMD
Mean age, years	41.3 ± 10.0	49.0 ± 10.6	-0.74	45.7 ± 10.2	46.2 ± 11.4	-0.042
Male (%)	58 (45.7)	129 (54.9)	-0.18	60 (48.5)	120 (51.0)	-0.049
Mean BMI, kg/m²	22.8 ± 1.1	22.3 ± 1.2	0.42	22.5 ± 1.1	22.5 ± 1.2	0.045
ASA I (%)	83 (65.4)	167 (71.1)	-0.12	90 (73.1)	166 (70.3)	0.063
Smoking history (%)	19 (15.0)	21 (8.9)	0.19	13 (10.3)	24 (10.2)	0.0043
Drinking alcohol (%)	29 (22.8)	40 (17.0)	0.15	20 (16.6)	42 (17.8)	−0.033
Mandible implant (%)	68 (53.5)	147 (62.6)	-0.18	74 (60.5)	143 (60.5)	-0.00057

Pain outcomes

Table 2 presents pain intensity scores at the three assessment time points. The loxoprofen group reported lower pain scores before surgeries. The most pronounced difference was observed on the evening of implant surgery. After 24 h of surgery, the loxoprofen group maintained significantly lower pain scores.

Table 2.

Pain intensity (VAS 0–10): IPTW-adjusted analyses

Time Point	Loxoprofen group	Control group	Adjusted Difference (95% CI)	p-value
30 min preoperative	1.64 ± 0.12	2.20 ± 0.10	-0.56 (-0.83, -0.30)	< 0.001
Evening of surgery	3.91 ± 0.12	5.70 ± 0.10	-1.80 (-2.06, -1.53)	< 0.001
24 h postoperative	2.36 ± 0.12	3.99 ± 0.10	-1.64 (-1.90, -1.37)	< 0.001

*Adjusted for all covariates listed in Table 1 using IPTW-weighted linear mixed-effects regression. Group×Time interaction p < 0.001, indicating that the treatment effect varied across time points. P-values were adjusted for multiple comparisons using the Tukey–Kramer method

We calculated the number needed to treat (NNT) to prevent one patient from experiencing moderate-to-severe pain (VAS ≥ 4). NNT for prevention of moderate-to-severe pain (VAS ≥ 4) after IPTW weighting was 24.89 (14.32, 76.69), 2.57 (1.98, 3.59), 2.18 (1.87, 2.64) 30 min preoperative, on the evening of surgery, and 24 h postoperative. NNT was calculated as 1/absolute risk reduction. NNT estimates were derived from IPTW-weighted logistic regression models, and CIs for NNT were calculated using the non-parametric bootstrap method.

Anxiety outcomes

The mean baseline T-AI score was 42.1 ± 0.77 in the loxoprofen group and 42.9 ± 0.69 in the control group, with no significant difference after IPTW adjustment (aMD − 0.79, 95% CI -2.74–1.16, p = 0.426). Table 3 presents S-AI scores for patients in the two groups across the three perioperative time points, adjusted for baseline T-AI scores using IPTW-weighted linear regression models. The loxoprofen group demonstrated significantly lower preoperative state anxiety scores, which persisted on the evening of surgery. The reduction in S-AI scores was attenuated at 24 h postoperatively.

Table 3.

Anxiety levels (S-AI 20–80) at three perioperative time points: IPTW-adjusted analyses with T-AI as a covariate

Time Point	Loxoprofen group	Control group	Adjusted Difference (95% CI)	p-value
30 min preoperative	47.82 ± 0.90	53.01 ± 0.72	-5.19 (-7.11, -3.27)	< 0.001
Evening of surgery	39.28 ± 0.90	47.86 ± 0.72	-8.57(-10.49, -6.65)	< 0.001
24 h postoperative	37.08 ± 0.90	39.32 ± 0.72	-2.25 (-4.17, -0.33)	0.022

* Adjusted for all covariates listed in Table 1 and T-AI using IPTW-weighted linear mixed-effects regression. Group×Time interaction p < 0.001, indicating that the treatment effect varied across time points. P-values were adjusted for multiple comparisons using the Tukey–Kramer method

NNT to prevent one patient from experiencing moderate-to-severe state anxiety (S-AI score ≥ 45) at each perioperative time point was 6.14 (3.66, 17.71), 2.54 (2.03, 3.63), and 12.66 (5.87, 124.75).

Postoperative infection and patient satisfaction

Adjusted for all covariates listed in Table 1 using IPTW-weighted logistic regression, five patients (3.9%) in the loxoprofen group and 13 patients (5.5%) in the control group developed infection after implant surgery (adjusted OR: 0.49 (0.14–1.74), p = 0.266). The difference between the two groups was no longer significant.

Overall patient satisfaction was significantly higher in the loxoprofen group (4.69 ± 0.059 VS 4.21 ± 0.069; aMD 0.48, 95% CI 0.29–0.67, p < 0.001). In the loxoprofen group, 94 patients (74.0%) rated their satisfaction as “very satisfied” (score 5), compared to only 127 patients (54.0%) in the control group (p < 0.001). Figure 2 shows the distribution of patient satisfaction levels on the seventh day postoperatively.

Fig. 2.

Distribution of patient satisfaction levels at the seventh day postoperatively

During the study, only three patients of the loxoprofen group (2.4%) reported mild gastrointestinal discomfort. Two of them reported nausea after taking loxoprofen sodium, and one patient complained about dyspepsia. One patient of the control group (0.4%) reported nausea after implant surgery. All the gastrointestinal adverse events were resolved spontaneously without medical intervention. No patients in the study reported symptoms suggestive of gastrointestinal bleeding, allergic reactions, cardiovascular events, liver adverse events, or kidney adverse events. The incidence of adverse events did not differ significantly between groups (p = 0.12 by Fisher’s exact test). PCA was performed to visualize the underlying structure of the clinical outcome variables (Supplementary Fig. 2). An IPTW-weighted radar chart was constructed comparing the two groups (Supplementary Fig. 3).

Subgroup analysis

Forest plots of subgroup analyses for VAS, S-AI, T-AI, and satisfaction scores are presented in Supplementary Figs. 4–6. The observed treatment effects showed a similar direction across most subgroups, with the exception of the smoking and ASA II subgroups for preoperative VAS, smoking subgroup for preoperative S-AI, where the differences did not reach statistical significance (p > 0.05). At 24 h postoperatively, the reduction of anxiety scores was attenuated but remained significant in some subgroups. The reduction was driven by female patients, patients with BMI < 22 kg/m², ASA I patients, smoking patients, non-drinkers, and patients experienced maxilla implants. The significant improvement in patient satisfaction in the loxoprofen group was observed across all subgroups except the smoking subgroup (p = 0.055).

The subgroup analyses were exploratory and were not powered to detect general differences within individual subgroups. The small sample sizes in certain subgroups limit the precision of results and the ability to draw definitive conclusions.

Interaction analysis and doubly robust sensitivity analysis

To further evaluate the consistency of treatment effects across patient subgroups, we tested interactions between loxoprofen sodium treatment and each baseline covariate. The p-values of interaction terms are summarized in Supplementary Table 1. Given the low incidence of postoperative infection, this study did not conduct a formal subgroup interaction analysis for the outcome.

For VAS pain scores, none of the interaction terms reached statistical significance across the three time points (p > 0.05), indicating that the analgesic effect of loxoprofen sodium was not significantly modified by age, sex, BMI, ASA status, smoking, drinking, or implant site. For S-AI, most interaction p-values were more than 0.05, except for BMI at 24 h postoperatively (p = 0.032), suggesting a potential differential effect by BMI at this time point. Specifically, the pain-relieving effect of loxoprofen sodium is stronger in patients with a BMI below 22, consistent with the subgroup result mentioned above. For T-AI, no significant interactions were detected, consistent with the absence of an overall treatment effect on this stable personality trait. Regarding patient satisfaction, all interaction p-values were greater than 0.05, indicating that satisfaction improved regardless of patients’ characteristics.

The analysis of analgesic and anxiety-reducing effects of loxoprofen sodium inherently adopted a doubly robust approach. We used IPTW to balance baseline covariates and simultaneously adjusted for them in subsequent linear mixed-effects models. Therefore, the estimated treatment effects are doubly robust.

Discussion

This retrospective cohort study using IPTW provides the first large-scale evidence evaluating the efficacy of loxoprofen sodium in dental implant surgery. IPTW is a robust approach to control confounding variables in observational studies, superior to simple multivariable adjustment or even traditional matching [35, 36]. Before weighting, the loxoprofen group was notably younger than the control group. It reflects the real prescribing pattern that surgeons prefer to prescribe analgesics to younger patients [37, 38]. By weighting patients to create a pseudo-population with balanced covariates, IPTW mimics the conditions of a randomized trial. Based on IPTW, our analysis yields several principal insights.

First, perioperative pain must be managed in implant surgery, and some medications, including acetaminophen, pregabalin and clonazepam, have been proven to be effective in reducing pain [39, 40]. NSAIDs are effective for postoperative dental pain relief, but there were very few studies conducted throughout the perioperative period [41–43]. Perioperative loxoprofen sodium provides superior and clinically meaningful pain control throughout the perioperative implant timeline. The statistically significant reduction in VAS scores, particularly on the evening of surgeries, is noteworthy. The loxoprofen group reported lower pain scores even before surgery, suggesting that the preoperative loxoprofen dose produced a measurable analgesic effect within 30 min of administration. This finding aligns with previous clinical trials demonstrating the efficacy of NSAIDs for the management of dental postsurgical pain [44, 45]. The observed analgesic profile aligns with the drug’s pharmacokinetic properties, with the preoperative dose effectively covering the immediate postoperative phase and subsequent doses maintaining control during the peak inflammatory window (several hours post-surgery). This specific pattern validates the three-dose regimen employed and underscores the importance of analgesic use before implant surgery, which has been overlooked in the current clinical practice of China. As uncontrolled night pain may disrupt sleep and increase rescue analgesic consumption [46], this level of pain control can directly translate into improved early postoperative quality of life. The NNT was between 2 and 3 on the evening of surgery and 24 h after surgery, indicating that only 3 patients needed to receive loxoprofen sodium to prevent 1 case of substantial postoperative pain.

Second, dental anxiety can lead to avoidance of dental treatment and impaired oral health, but pharmacological options are often limited by side effects and cost [17, 47]. The use of non-drug treatments, like cognitive-behavioural therapy intervention, was effective in reducing dental anxiety [48]. Loxoprofen sodium administration is associated with significantly lower preoperative and early postoperative state anxiety. The most pronounced anxiolytic effect was observed on the evening of surgery, indicating that treating only three patients with loxoprofen sodium could prevent one patient from experiencing substantial anxiety. This observation extends the drug’s known therapeutic profile. While this observation is clinically important, the underlying mechanisms warrant cautious interpretation. The preoperative dose may have provided patients with a sense of having their pain managed, which reduced anticipatory anxiety and indirectly lowered anxiety [49–51]. Inflammation is also linked to mood disturbances [52, 53]. Loxoprofen sodium may exert an anxiolytic effect by exerting the anti-inflammatory effect. However, this mechanism remains hypothetical and requires investigation in future studies. The subgroup analyses further strengthened the robustness of the findings above. The analgesic effect of loxoprofen sodium was observed across all clinically relevant subgroups. The anxiolytic effect on state anxiety also showed remarkable consistency across subgroups before and on the evening of implant surgery. However, these subgroup analyses were exploratory and underpowered for certain subgroup comparisons. No significant treatment effect for T-AI was observed, indicating loxoprofen sodium specifically affects S-AI rather than T-AI. This is understandable, since loxoprofen sodium belongs to NSAIDs rather than medications treating anxiety disorders.

Next, no decrease in postoperative infection was detected. The significant improvement in satisfaction associated with loxoprofen sodium was observed across all clinically relevant subgroups, a trend increasingly recognized as a core quality metric in implant dentistry [54]. The finding may reflect the combined benefits of better pain control and lower anxiety rather than a direct effect of loxoprofen sodium itself. Future studies should assess whether this magnitude of difference translates into clinically meaningful improvements.

The absence of statistical significance in the interaction analysis underscores the generalizability of our findings. One notable exception was the interaction between treatment and BMI on S-AI 24 h after implant surgery (p = 0.032). However, this finding could be due to the chance of the limited number of patients, and further investigation with larger participants is needed in future studies.

However, several limitations should be acknowledged in the study. First, though we used rigorous propensity score weighting, residual confounding from unmeasured variables, including preoperative pain threshold, baseline dental anxiety, socioeconomic status, implant system, bone density, implant site details, and surgeon effect, cannot be entirely excluded. Second, all implant surgeries in the study were performed at a single hospital in China, which limits its generalizability to other populations. Third, the data only recorded the situation 7 days after implant surgery and is insufficient to assess late-onset postoperative infection like peri-implant mucositis or peri-implantitis, which require longer-term clinical and radiographic monitoring. Fourth, the extensive analyses performed increased the risk of type I error due to multiple comparisons. Analyses such as subgroup analyses, interaction tests, and NNT calculations were exploratory and not adjusted for multiplicity. Therefore, these findings should be interpreted cautiously. In future, multicenter randomized controlled trials with extended follow-up are warranted to validate our findings across different surgical teams.

Conclusions

This cohort study provides the first large-scale evidence that a standardized perioperative regimen of loxoprofen sodium confers significant, multifaceted benefits for patients undergoing single-tooth dental implant surgery. Loxoprofen sodium significantly reduces perioperative pain intensity, with the greatest effect observed on the evening of surgeries. The usage of the drug is associated with significantly lower preoperative and early postoperative anxiety scores. No decrease in the risk of postoperative infection was detected, and overall patient satisfaction is significantly improved in the loxoprofen sodium group. The beneficial effects of loxoprofen sodium appeared generally consistent across the patient subgroups examined.

Loxoprofen sodium may represent an effective, inexpensive, and well-tolerated analgesic option for implant surgery in China. The observed reduction in anxiety scores in the loxoprofen group may reflect indirect effects related to pain relief, and more mechanistic studies are needed to elucidate the association between anxiety level and the drug. Future research should focus on validating these findings in definitive multicenter randomized controlled trials with an extended follow-up period and exploring potential central mechanisms of the anxiolytic effects of loxoprofen sodium.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

AL contributed to conception and design, data acquisition, data analysis, drafted and critically revised the manuscript. HP contributed to data analysis and interpretation, and drafted the manuscript. DL and RH contributed to data acquisition, and critically revised the manuscript. ML contributed to conception and design, and critically revised the manuscript. All authors gave final approval of the manuscript, and agreed to be accountable for all aspects of work ensuring integrity and accuracy.

Funding

This work was supported by the Changsha Natural Science Foundation Project [grant number: kq2502309], the Undergraduate Research and Innovation Fund Program of Hunan University of Chinese Medicine [grant number:2024BKS034]; and the General Project of University and College Joint Fund of Hunan University of Chinese Medicine [grant number: 2024XYLH298].

Data availability

The datasets generated and analyzed during the current study are not publicly available due to the privacy policy of Changsha Stomatological Hospital, but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by Medical ethics committee of Changsha Stomatological Hospital. The ethics approval number was (2024) Ethical review [research] No.88, and the date of approval was March 1st, 2024. Informed consent forms were signed for patients participated in the study.

Competing interests

The authors declare no competing interests.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process

During the preparation of this work the authors used Deepseek in order to edit the language and modify the R language code used for figure drawing. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article.