Peri-Implantitis Causal Therapy with and Without Doxycycline: Retrospective Cohort Clinical Study

Abstract

Background: Topical application within peri-implant pockets ensures high drug concentrations at the infection site while minimizing systemic exposure. However, the comparative effectiveness of non-surgical causal therapy alone versus its combination with doxycycline remains unclear. This retrospective observational clinical study aimed to evaluate the impact of adjunctive doxycycline on peri-implant parameters, considering smoking, systemic conditions, and implant–prosthetic rehabilitation (single implant, implant-supported bridge, or full-arch). Methods: Patients were retrospectively assigned to a control group (CG), receiving non-surgical causal therapy alone, or a test group (TG), which is also treated with topical doxycycline. Peri-implant parameters, including Peri-implant Probing Depht (PPD), Bleeding on Probing (BoP), Plaque Index (PI), and suppuration, were assessed at baseline (T0) and follow-up (T1). Multivariate logistic regression and stratified subgroup analyses were conducted to adjust for confounders such as smoking, systemic conditions, and implant–prosthetic rehabilitation types. Results: Two hundred nine patients were included in the study, of whom 97 were in the CG and 112 were in the TG. At T1, the TG exhibited a statistically significant reduction in PPD, BoP, PI, and suppuration compared to the CG (p < 0.05). Conclusions: The adjunctive use of topical doxycycline significantly enhances clinical outcomes in non-surgical peri-implantitis treatment. Further longitudinal studies are needed to confirm these findings and assess long-term stability.

1. Introduction

Dental implants represent the gold standard in the rehabilitation of single, partial, or fully edentulous patients, reporting a high medium- and long-term success rate [1,2,3,4].

However, several studies report a high incidence of peri-implantitis 5–10 years after implant placement [5,6,7,8,9].

Peri-implantitis represents an inflammatory pathological condition involving soft and hard tissues around dental implants, which is marked by progressive bone loss due to the bacterial biofilm build-up and related chronic inflammation [10,11].

In contrast to peri-implant mucositis, which involves only soft tissues without resulting in bone loss, peri-implantitis represents a more critical and potentially compromising clinical condition for implant fixture stability [12,13].

The occurrence of peri-implantitis is impacted by several factors. Known risk indicators include individual behaviors and conditions such as smoking and diabetes mellitus, which impair immune response and the healing process, resulting in patients being more susceptible to chronic bacterial infection [14]. In addition, inadequate oral hygiene is a key factor in peri-implantitis development, as it allows bacterial biofilm aggregation around the implant. Other risk determinants include genetic predisposition (e.g., the presence of specific alleles for pro-inflammatory cytokines such as IL-1β) and local factors such as implant geometry, prosthetic restoration design, and implant placement. Implants with rough surfaces or a subcrestal location increase the likelihood of deep bacterial colonization, making decontamination more difficult [15,16].

The therapeutic management of peri-implantitis is multifaceted and includes non-surgical and surgical approaches, each with distinct goals. Non-surgical therapies aim to decontaminate implant surfaces using mechanical instruments, such as titanium scalers, sonic instruments, and ultrasound with compatible material inserts. However, effective removal of bacterial biofilm could be impaired by the roughness of the implant surface and threads, where biofilm may persist even after treatment [17]. Chemical approaches, such as chlorhexidine or hydrogen peroxide injection to irrigate the peri-implant pocket, are often combined with mechanical decontamination to increase the effectiveness of bacterial removal [18]. Although these strategies are available, mechanical decontamination alone may not be enough to halt peri-implantitis progression in severe cases, requiring additional adjuvant interventions.

Antibiotic employment in the management of peri-implantitis has been investigated in several clinical investigations and systematic reviews. Antibiotics can be administered systemically or locally.

Systemic antibiotics, such as amoxicillin and metronidazole, are often recommended for reducing bacterial load in the context of severe infections. However, systematic reviews by Esposito et al. [19] and Polymeri et al. [20] suggest that systemic antibiotics are not the most effective solution in the peri-implantitis management context. Systemic antibiotics fail to achieve adequate concentrations in peri-implant pockets due to the presence of bacterial biofilm, which acts as a protective barrier. In addition, the systemic use of antibiotics raises concerns about the increased risk of bacterial resistance development and systemic side effects, such as allergic reactions and gastrointestinal disorders [21]. For these reasons, the efficacy of systemic antibiotics in managing peri-implantitis is often temporary, with short-term clinical improvement not resulting in long-term prevention of bone loss.

Local antibiotics represent a therapeutic option that has gained increasing relevance. The application of antibiotics directly into the peri-implant pocket allows for high concentrations of the drug at infected sites, improving clinical efficacy and minimizing systemic side effects. Recent studies have shown that the use of doxycycline or minocycline in the form of gels or microspheres applied locally in the peri-implant pocket, in combination with mechanical decontamination, leads to a significant reduction in probing depth and bleeding levels compared with mechanical decontamination alone [22]. Toledano et al., in a 2021 meta-analysis, confirmed that the use of local antibiotics led to a clinically significant reduction in bacterial load and clinical signs of inflammation [23]. Grusovin et al. also reported improvements in clinical parameters with the use of topical antibiotics, suggesting that this treatment modality may optimize the results of nonsurgical therapy and prevent the need for surgery [24].

Such topical formulations, particularly doxycycline, are often delivered in the form of bioabsorbable gels or microencapsulated microspheres that provide sustained release of the active ingredient for a period generally ranging from 7 to 14 days. After application at the infected site, doxycycline reaches high local concentrations, exceeding the minimum inhibitory concentration (MIC) for the main periodontal and peri-implant pathogens. The mechanism of action is based on the inhibition of bacterial protein synthesis through binding to the 30S ribosomal subunit. In addition, doxycycline also possesses documented anti-inflammatory properties, including the inhibition of matrix metalloproteinases (MMPs), which contribute to the reduction of soft and hard tissue destruction. This dual effect, antibacterial and modulating the host response, makes topical formulations a valuable adjunct to non-surgical therapy in cases of incipient mucositis or peri-implantitis [25,26].

The aim of this retrospective observational clinical study was to compare non-surgical causal peri-implant therapy with and without doxycycline according to variables smoking, systemic diseases, and a kind of implant–prosthetic rehabilitation (single, bridge, or full arch).

2. Materials and Methods

2.1. Study Design

The current retrospective observational clinical cohort study was conducted at the Department of Dentistry and Dental Prosthetics, IRCCS Ospedale San Raffaele, Milan, Italy. The investigation was in accordance with the principles stated in the Declaration of Helsinki and was approved by the Ethics Committee (approval number: 190/INT/2021). The conduct of the survey followed STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines, which can be found at the following link: http://www.strobe-statement.org/ (accessed in date 4 January 2024).

2.2. Setting

This retrospective clinical study analyzed patients who underwent non-surgical peri-implant therapy before and after the introduction of 14% topical doxycycline gel (Ligosan^®) at the study center. To ensure methodological consistency and minimize temporal biases, patient selection was stratified according to treatment periods as follows:

• Control Group (CG): This cohort included patients who received non-surgical peri-implant therapy between January 2021 and December 2022 before the introduction of Ligosan^® in clinical practice. These individuals underwent a standardized non-surgical treatment protocol consisting of professional mechanical debridement without local or systemic antibiotic drugs.
• Test Group (TG): This cohort comprised patients who underwent the same non-surgical peri-implant therapy between January 2022 and March 2024, after the introduction of Ligosan^®. In addition to mechanical debridement, these patients were treated with 14% topical doxycycline gel to assess its adjunctive efficacy in peri-implant disease management.

Patients were retrospectively selected through clinical records, ensuring that all clinical procedures, data collection, and follow-up evaluations adhered to the same standardized protocol across both groups. This study design allowed for a direct comparison of peri-implant clinical outcomes before and after the implementation of topical doxycycline, providing insights into its therapeutic impact on peri-implant health.

The protocol procedures, including data recording, were performed by three operators (dentists) with more than 5 years of clinical experience in periodontology and implantology.

2.3. Participants

Patients included in the study had undergone a professional oral hygiene session or attended a first periodontal visit during which periodontal screening and recording (PSR) indicated a score of 3 or 4, prompting further diagnostic evaluation for peri-implant disease. Only patients aged ≥18 years with a confirmed diagnosis of peri-implantitis, classified according to the 2018 Classification of Periodontal and Peri-Implant Diseases and Conditions proposed by Chapple et al. [27], were included.

2.4. Inclusion Criteria

Patients were eligible for inclusion if they met the following criteria:

• Age ≥ 18 years;
• Presence of at least one osseointegrated implant diagnosed with peri-implantitis based on radiographic and clinical criteria, according to the 2018 Classification of Periodontal and Peri-Implant Diseases and Conditions [15];
• Subjected to exclusively screw-retained full-ceramic implant–prosthetic rehabilitation, regardless of implant design;
• Availability of complete clinical records, including baseline (T0) and follow-up (T1) evaluations;
• Patients with compensated systemic conditions (e.g., controlled diabetes or cardiovascular disease);
• Smokers were not excluded to reflect real-world clinical conditions and assess the impact of smoking on treatment outcomes.
• Exclusion criteria
• Patients were excluded if they met any of the following conditions:
• Lack of complete clinical documentation, including missing baseline or follow-up data;
• Previous history of surgical peri-implant therapy within 12 months prior to study inclusion;
• Use of local or systemic antibiotics for peri-implant treatment within the 3 months preceding non-surgical therapy;
• Presence of autoimmune diseases or uncompensated systemic conditions (e.g., uncontrolled diabetes, immunosuppression) that could interfere with tissue healing and disease progression after treatment;
• Inability to comply with the scheduled follow-up protocol, due to either non-compliance or logistical constraints;
• Participation in concurrent clinical trials evaluating peri-implant disease management during the study period.

According to eligibility criteria, a total of 209 patients, of whom 97 were enrolled in CG and 112 in TG, were considered for the study.

2.5. Variables

The following parameters were collected for each participant: age, smoking, systemic diseases, and kind of implant–prosthetic rehabilitation (single implants, bridge-supported implants, and full-arch implant–prosthetic rehabilitations).

Peri-implant clinical indexes, including Placque Index (PI), Bleeding on Probing (BoP), Peri-implant Probing Depth (PPD), and suppuration, were evaluated at T0 and 3 months after treatment conclusion (T1).

A periodontal chart was completed for each patient in T0 and T1. A calibrated periodontal probe (UNC-15, Hu-Friedy, Chicago, IL, USA) was employed and inserted for each parameter at six points: mesial, medial, and distal on both the buccal and lingual/palatal sides.

The PI was determined through the application of a plaque detector. The value was expressed as a percentage (0–100%), indicating the proportion of implant surfaces with visible bacterial biofilm.

BoP was assessed 30 s after removal of the probe and noted at all six probing sites for each implant. The BoP value was reported as the percentage of sites with bleeding (0–100%), providing an indicator of peri-implant tissue inflammation.

The PPD measured the vertical distance between the peri-implant mucosal margin and the base of the pocket in each implant. This measurement was also taken at six sites per implant, and the values were expressed in millimetres (mm).

Suppuration was detected by probing peri-implant sulcus and observing presence or absence of purulent exudate, whether spontaneous or induced. The suppuration parameter was recorded as a dichotomous variable (present/absent).

2.6. Non-Surgical Treatment Protocol (CG)

Two sessions were performed involving the use of piezoelectric ultrasonic tips, Electro Medical Systems (E.M.S), Nyon, Switzerland screwed into a Piezon handpiece associated with a Piezon 250 device with regulated power (Figure 1).

Piezoelectric technology allowed high-frequency linear ultrasonic vibrations, specifically between 28,000 and 32,000 Hz.

The initial debridement was conducted using the A tip, specifically designed for supragingival calculus removal. This was followed by Perio (P) tip, which provided removal of both supragingival and subgingival deposits, reaching depths of up to 3 mm. For mechanical disruption of biofilm on implant surfaces affected by marginal bone loss, the Perio Implant (PI) tip, featuring non-abrasive properties, was employed to achieve thorough biofilm removal while preserving implant surface’s structural integrity (Figure 2).

2.7. Non-Surgical Treatment Protocol Applying Doxycycline (TG)

The same two sessions as CG were performed in TG. At the end of the second one, topical doxycycline gel 14% (140 mg/g) (Ligosan, Kulzer, Hanau, Germany) was applied with dedicated dispenser (Figure 3).

The agent was introduced into the peri-implant pocket through slim applicator; any overflow was pushed into the site by periodontal probe. The leftover antibiotic was bagged and provided to the patient so that it could be stored in the refrigerator, brought back at the next session (10 days later), and again placed in the site.

In both observed groups, a new periodontal chart was completed 3 months later.

2.8. Data Sources

Considering kind of implant–prosthetic rehabilitation, systemic diseases, and smoking as variables, the difference in the development of peri-implant clinical parameters from T0 to T1 in GC and GT was evaluated.

2.9. Bias

Selection bias was minimized by including all eligible patients treated within predefined, non-overlapping timeframes, ensuring that treatment allocation was determined solely by the introduction of 14% topical doxycycline gel (Ligosan^®) rather than patient-specific factors. Information bias was controlled by extracting clinical data from standardized electronic health records (EHRs) and structured periodontal charts, with peri-implant parameters systematically recorded at six sites per implant (mesial, medial, and distal on both buccal and lingual/palatal surfaces). Measurement bias was mitigated through examiner calibration sessions, ensuring high intra- and inter-examiner reliability (ICC ≥ 0.85), the use of a UNC-15 periodontal probe (Hu-Friedy, Chicago, IL, USA) with a standardized probing force (≤25 g), and duplicate measurements at each site to enhance reproducibility.

Confounding bias was addressed using multivariate logistic regression models and stratified subgroup analyses, adjusting for potential confounders such as smoking status, systemic conditions, and implant–prosthetic rehabilitation type. Temporal bias was minimized by implementing distinct treatment periods for CG (January 2021–December 2022) and TG (January 2022–March 2024), ensuring that treatment protocols were identical except for the adjunctive application of topical doxycycline gel in TG. Although observer bias was inherent due to the retrospective nature of the study, follow-up evaluations were conducted by independent examiners blinded to the initial treatment allocation, ensuring objectivity in clinical outcome assessment.

2.10. Study Size Rationale

The study sample size was determined through a priori statistical power analysis, setting the significance level (α) at 0.05 and the effect size at 0.5, to ensure an optimal balance between Type I and Type II error rates. The calculation was performed using the following formula:

$$n=\left(2\times \left(Za2+Zp\right)2\times o2\right)IES2$$

Based on this analysis, a total of 209 participants (97 in the control group and 112 in the test group) were included, yielding a statistical power of 0.97. This level of power ensures a high probability of detecting clinically significant differences between groups, reducing the likelihood of Type II errors. The final sample size exceeded the minimum requirement for adequate inferential strength.

2.11. Statistical Methods

Statistical analysis was conducted using IBM SPSS Statistics software, version 29, to assess changes in peri-implant parameters CG and TG. The parameters analyzed included PPD, BoP, PI, and suppuration, with a focus on three main variables: systemic diseases, smoking habits, and the kind of implant–prosthetic rehabilitation (single, bridge-supported implants, or full-arch rehabilitation).

Variables were divided into two categories: continuous and categorical. The continuous variables included clinical parameters, i.e., PPD, BoP, and PI, measured as numerical values to represent specific variations in peri-implant conditions. Suppuration, which was initially categorical, was converted to a numeric binary variable (0 = no suppuration, 1 = presence of suppuration) to enable statistical analysis. Categorical variables included presence or absence of systemic diseases, smoking habit (smokers or non-smokers), and type of implant–prosthetic rehabilitation (classified as single, bridge-supported implants, and full arch).

For each parameter, a null hypothesis was formulated. For PPD, the null hypothesis was that there were no significant differences between CG and TG in the parameter changes (p > 0.05). Similarly, for the BoP, it was assumed that there were no significant differences between the two groups (p > 0.05). For PI, the null hypothesis stated that there was no significant change in the value between CG and TG from T0 to T1 (p > 0.05). For suppuration, the null hypothesis assumed that there was no significant difference in the incidence of suppuration between the evaluated groups (p > 0.05).

Statistical analyses for the comparison between CG and TG were performed using Student’s t-test for independent samples. The relationship between the continuous and categorical variables was evaluated through Pearson’s correlation coefficients for continuous variables, while the Chi-square test and ANOVA were applied for categorical variables. For each of these analyses, the null hypothesis predicted the absence of significant correlations between the independent variables (systemic diseases, smoking, type of rehabilitation) and the peri-implant parameters analyzed (p > 0.05).

The data were preliminarily processed to ensure standardization of the categorical variables and correct handling of any missing values. Analyses were conducted sequentially, respecting a level of statistical significance set at p < 0.05.

3. Results

Two hundred nine patients were included in the study, of whom 97 were in the CG and 112 were in the TG.

One patient in CG and four in TG were included in the dropout due to not having complied with the recalls in the timeframe provided by the protocol. Sample features were reported as follows (

Table 1. Sample retained for study purposes.

	CG	TG
Number of patients	96	108
Average age (Range)	55 (38–72)	54.5 (34–75)
Female	37	45
Males	59	63
Affected by systemic diseases and no smokers	19	23
Affected by systemic diseases and smokers	7	21
Healthy patients and no smokers	55	46
Healthy patients and smokers	15	18
Single implant	43	48
Bridge-supported implants	16	19
Full-arch implant–prosthetic rehabilitation	37	41

).

3.1. PPD

In the PPD, the null hypothesis was that there was no significant difference between measurements at time 0 and after follow-up within the groups. This hypothesis was rejected for both CG and TG (p < 0.05). In the CG, the mean PPD decreased from 10.39 mm at time 0 to 9.20 mm at the end of follow-up, indicating a non-significant reduction (p > 0.05). In contrast, in TG, the mean PPD decreased significantly from 7.51 mm to 5.30 mm (p < 0.05) (Figure 4).

In subjects with systemic diseases, the PPD decreased significantly in the TG (from 7.53 mm to 5.40 mm; p < 0.05) compared to the CG (from 10.67 mm to 9.50 mm; p > 0.05). Similar results were observed in non-smokers and smokers, with a greater reduction in TG than in CG. Analyzing the kind of rehabilitation, the TG showed significant reductions in PPD for single (from 6.30 mm to 4.50 mm; p < 0.05), bridge-supported (from 7.00 mm to 5.20 mm; p < 0.05), and full-arch (from 7.38 mm to 5.40 mm; p < 0.05) rehabilitations, while in the CG, the reductions were not significant.

3.2. BoP

For BoP, the null hypothesis predicted no significant change between time 0 and follow-up. This hypothesis was rejected for TG (p < 0.05) but confirmed for CG (p > 0.05).

In the CG, the mean BoP decreased non-significantly (from 0.40 to 0.35; p > 0.05), whereas in the TG, a significant decrease was observed (from 0.25 to 0.10; p < 0.05) (Figure 5).

In subjects with systemic disease, the TG showed a significant decrease in BoP (0.30 to 0.15; p < 0.05), whereas no significant changes were observed in the CG. In non-smokers, BoP decreased significantly in the TG (p < 0.05), whereas a smaller reduction was observed in smokers. Concerning the type of rehabilitation, significant reductions in BoP were observed in the TG for single (0.20 to 0.05; p < 0.05), bridge-supported (0.25 to 0.10; p < 0.05), and full-arch (0.30 to 0.15; p < 0.05) rehabilitations, whereas in the CG, the changes were non-significant.

3.3. PI

The null hypothesis, which predicted no change in the PI, was rejected in the TG (p < 0.05) but confirmed in the CG (p > 0.05). In the CG, the mean PI increased from 0.45 to 0.40 without statistical significance, whereas in the TG, it decreased significantly from 0.30 to 0.15 (p < 0.05) (Figure 6).

In patients with systemic diseases, the TG showed a significant decrease in PI (from 0.35 to 0.20; p < 0.05), in contrast to the CG (p > 0.05). In non-smokers, the PI decreased significantly in the TG (p < 0.05), whereas in smokers, the improvement was less evident. According to the type of rehabilitation, the TG showed significant reductions in single rehabilitation (0.25 to 0.10; p < 0.05), supporting bridges (0.30 to 0.15; p < 0.05), and the full arch (0.35 to 0.20; p < 0.05), whereas no significant differences were observed in the CG.

3.4. Suppuration

For suppuration, the null hypothesis was rejected in the TG (p < 0.05) but confirmed in the CG (p > 0.05). In the TG, the incidence of suppuration decreased significantly from time 0 to follow-up, from 0.20 to 0.05 (p < 0.05), whereas no significant change was observed in the CG (0.35 to 0.30; p > 0.05) (Figure 7).

The reduction in TG was significant in all subgroups analysed, including patients with systemic diseases, smokers, and non-smokers. For the type of rehabilitation, the TG showed significant reductions in suppuration in single (0.10 to 0; p < 0.05), bridge-supported (0.15 to 0.05; p < 0.05), and full-arch (0.20 to 0.05; p < 0.05) rehabilitations, whereas no statistically significant differences were found in the CG.

4. Discussion

The results of this retrospective study showed a significant improvement in peri-implant parameters in the group treated with non-surgical causal therapy combined with topical doxycycline compared to the group treated with non-surgical causal therapy alone. The improvement seen in TG in terms of reduction of PPD, BoP, PI, and suppuration confirms the efficacy of using local doxycycline as an adjunctive treatment for the management of peri-implantitis.

Similar results were reported by Ding et al., who evaluated, using a mouse model, the efficacy of applying doxycycline to implant surfaces to modulate the progression of peri-implantitis. The study used an experimental model in which implants treated with doxycycline were compared to untreated implants. The results showed a significant reduction in peri-implant bone loss (2.3 ± 0.4 mm in controls vs. 0.9 ± 0.2 mm in treated) and a decrease in the inflammatory infiltrate in the treated sites compared to the control group, supporting the efficacy of the antibiotic in improving local inflammatory parameters [28].

Nastri et al. conducted an in vitro study to evaluate the antimicrobial efficacy of a controlled release system containing metronidazole and doxycycline. The results showed a significant reduction in bacterial growth of Aggregatibacter Actinomycetemcomitans and Porphyromonas Gingivalis, two key pathogens in peri-implantitis, confirming the efficacy of sustained-release antibiotics [29].

A systematic review by Passarelli et al. analysed data from clinical and randomised controlled trials to evaluate the efficacy of local antibiotics in peri-implantitis management. The review included 15 studies with a total of 742 patients and reported an average PPD reduction of 1.5 mm and BoP improvement of 30% in patients treated with local antibiotics compared to mechanical decontamination alone, suggesting a significant clinical benefit of additional antibiotic therapy [30].

The randomized controlled trial by Bücher et al. evaluated the efficacy of doxycycline extended release in the treatment of peri-implantitis in a sample of 50 patients followed for 6 months. The doxycycline-treated group showed an average reduction in PPD of 2.1 mm compared to a reduction of 0.9 mm in the control group, with a significant improvement also in BoP (−35% in the treated group) [31]. Arshad et al. conducted a randomized split-mouth clinical trial of 30 patients evaluating the prophylactic application of doxycycline at the implant–abutment interface. After a 12-month follow-up, BoP was reduced by 40% in treated sites compared to control sites, suggesting a positive effect of the antibiotic in preventing peri-implant inflammation [32].

The issue of antibiotic resistance is a critical factor in the management of peri-implantitis. A systematic review conducted by Ardila et al. examined the phenomenon of resistance in patients with peri-implant infections, showing that 35% of Porphyromonas Gingivalis strains isolated from infected sites had resistance to commonly used antibiotics, including doxycycline and metronidazole, emphasising the need for prudent use of these therapeutic agents [33]. Also, van Winkelhoff, in a literature review, suggested a selective approach to the use of local antibiotics to minimise the risk of resistance and recommended their use only in cases of advanced peri-implantitis with documented presence of specific pathogens [34].

Smoking has been widely recognized as a risk factor for peri-implantitis, as well as for treatment effectiveness. Reis et al. conducted a meta-analysis of 18 clinical studies with a total of 3215 patients and reported that smokers had a peri-implantitis incidence of 22.1%, significantly higher than non-smokers (12.3%). Similarly, Sgolastra et al. showed that the risk of peri-implantitis in smokers is 2.5 times higher than in non-smokers, confirming the negative influence of smoking on implant prognosis [35,36]. Cheng further confirmed this association, suggesting that smoking patients should be subjected to more stringent monitoring protocols to prevent the progression of peri-implant disease [37].

In addition, systemic conditions can influence peri-implant health and treatments. D’Ambrosio et al. conducted a systematic review that included 45 studies and found that patients with uncontrolled diabetes had a 30 per cent increased risk of developing peri-implantitis compared to healthy patients [38]. Yan et al. and Chhina, in their systematic literature reviews, reported that peri-implantitis is associated with elevated levels of systemic inflammatory markers, such as C-reactive protein, suggesting a connection between systemic inflammation and peri-implant disease [12,39].

In the present study, patients with systemic disease showed a less marked reduction in PPD and BoP than healthy subjects, suggesting a negative impact of systemic conditions on treatment response. However, topical doxycycline demonstrated a positive effect in improving peri-implant parameters, confirming its role as an adjunct in the non-surgical treatment of peri-implantitis.

5. Conclusions

Within the limitations of this retrospective study, the adjunctive use of topical doxycycline in non-surgical peri-implantitis treatment resulted in significantly greater improvements in clinical parameters compared to mechanical therapy alone. Overall, the results showed that the test group (TG) achieved significant improvements in peri-implant parameters from baseline to follow-up, regardless of systemic condition, smoking status, or type of implant-supported rehabilitation.

These findings support the clinical benefit of combining topical doxycycline with mechanical debridement in the management of peri-implant disease. Further prospective studies are needed to confirm these results and evaluate long-term outcomes.