ABSTRACT
Introduction
The effects of self‐reported penicillin allergy and alternative antibiotic regimens on implant outcomes remain poorly investigated. Therefore, the aim of this study was to evaluate the association between self‐reported penicillin allergy and the use of alternative antibiotics with post‐implant placement surgical complications, early implant failure, and preloading crestal bone loss (PLCBL).
Methods
This retrospective cohort study examined dental records of patients who received at least one dental implant. Patients who reported a penicillin allergy were identified as the exposed group, whereas patients without a reported penicillin allergy were selected as unexposed in a 1:2 exposed‐to‐unexposed ratio. Relevant patient‐ and implant‐related factors were recorded. Implants were evaluated for postsurgical complications, PLCBL, and early failure. Data were analyzed using bivariate analyzes and mixed‐effects regression models.
Results
A total of 179 subjects and 374 implants were evaluated. Early implant failure and surgical complications occurred in 6.1% and 8.0% of implants, respectively, while PLCBL was observed in 32.5% of implants. A higher prevalence of surgical complications and implant failure was observed among patients who reported a penicillin allergy compared with those without a reported allergy (p < 0.05). Implants prescribed azithromycin showed a higher prevalence of surgical complications, and those prescribed azithromycin or clindamycin had higher early failure rates compared with penicillin (p < 0.05). Regression analyzes showed that self‐reported penicillin allergy was not independently associated with surgical complications, PLCBL, or early implant failure (p > 0.05). In exploratory multivariable analyzes, azithromycin and other alternative antibiotics were associated with higher odds of surgical complications compared with penicillin (p < 0.05). No significant associations were found between antibiotic type and PLCBL or early implant failure (p > 0.05).
Conclusion
Self‐reported penicillin allergy was not independently associated with early implant outcomes in this cohort. Certain alternative antibiotics were associated with a higher prevalence of early surgical complications, warranting further investigation in larger studies.
Keywords: antibiotics, dental implants, penicillin
1. Introduction
Dental implants have become the preferred option for tooth replacement due to their high long‐term success rates and promising patient‐reported outcomes [1]. Despite cumulative high survival rates over a 20‐year period, early and late implant‐related complications continue to occur and may compromise the success of implant‐supported rehabilitation [2, 3, 4]. These complications may impose additional financial burdens, undermine patient expectations, and adversely affect overall quality of life by compromising masticatory function. Early implant complications, such as preloading crestal bone loss (PLCBL), failure to achieve or maintain osseointegration, and postoperative infection, are clinically relevant outcomes with multifactorial etiologies and complex risk profiles [5, 6, 7, 8]. Thus, identifying the prevalence of these early events and the risk factors associated with their occurrence is essential to improve clinical decision‐making, optimize implant treatment planning, and minimize delays in definitive rehabilitation.
The use of pre‐ and/or postoperative antibiotics in oral surgery has been a common clinical practice to reduce the risk of infection, with penicillin being the most frequently prescribed antibiotic [9, 10, 11, 12, 13, 14]. Approximately 10% of the United States population reports a history of allergy to penicillin or other β‐lactam antibiotics; however, true penicillin hypersensitivity is uncommon, with less than 1% of individuals having a confirmed penicillin allergy. Despite this discrepancy, penicillin allergy labels often persist in medical records and substantially influence prescribing practices. Patients labeled as penicillin‐allergic are more likely to receive alternative antibiotics, a practice that has been associated with increased risks of infection, antimicrobial resistance, and adverse drug reactions [15, 16, 17, 18, 19, 20, 21].
Self‐reported penicillin allergy and the subsequent use of alternative antibiotics have been associated with an increased risk of surgical site infection following various medical and dental surgical procedures [15, 22, 23, 24]. In implant dentistry, where antibiotic use is routine, this reliance on alternative antibiotics may also influence clinical outcomes. Previous investigations have reported that patients with a self‐reported penicillin allergy, who are often prescribed alternative agents such as clindamycin, exhibit higher rates of implant failure and postoperative infection compared with nonallergic patients [24, 25, 26, 27, 28, 29, 30]. Furthermore, in our previous retrospective study evaluating multiple patient‐ and implant‐related risk factors, self‐reported penicillin allergy emerged as a significant risk factor for PLCBL [7].
Despite this emerging evidence, the influence of self‐reported penicillin allergy, as well as the potential differential effects of commonly prescribed alternative antibiotic regimens, on early implant outcomes remains insufficiently investigated. Therefore, the aim of this study was to evaluate the association between self‐reported penicillin allergy and the use of alternative antibiotics with post‐implant placement surgical complications, early implant failure, and PLCBL. A better understanding of these associations may help determine whether patients who report penicillin allergy would benefit from formal allergy delabeling through confirmatory testing or, alternatively, from the selection of more appropriate substitute antibiotics.
2. Material and Methods
2.1. Study Design, Patient Selection, and Exposure Group Definition
This university‐based retrospective cohort study was conducted using electronic health record (EHR) data from patients who underwent placement of at least one dental implant at the University of Florida College of Dentistry (Gainesville, Florida, USA) between January 2011 and July 2021. The study protocol was approved by the University of Florida Institutional Review Board (IRB#202101382). The conduct and reporting of this study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.
From the pool of eligible patient records, a subset was randomly selected for detailed chart and radiographic review due to the extensive data extraction required. This approach allowed for a manageable yet representative sample while minimizing selection bias. Within this subset, patients who met the inclusion criteria and reported a penicillin allergy were classified as the exposed group. For each exposed patient, two eligible patients who met the same inclusion criteria but did not report a penicillin allergy were selected from the same source population and study period as the unexposed comparison group, resulting in a 1:2 exposed‐to‐unexposed ratio. This matching strategy was adopted to improve comparability between groups and minimize selection bias, rather than to estimate the true prevalence of self‐reported penicillin allergy in the studied population.
2.2. Inclusion Criteria
Patients were eligible for inclusion if they (1) underwent placement of at least one dental implant at the University of Florida Dental Clinics and (2) had complete clinical and radiographic documentation from implant placement through implant uncovering/abutment installation or prosthesis delivery. Patients were eligible regardless of antibiotic prescription status; however, when systemic antibiotics were prescribed in association with implant placement, they should be administered either preoperatively with postoperative continuation or exclusively in the postoperative period.
2.3. Exclusion Criteria
Patients/implants were excluded if clinical or radiographic data required to assess the study outcomes were incomplete or unavailable, including a lack of post‐implant placement follow‐up. Participants/implants were also excluded if radiographic quality (e.g., distortion, superimposition, or improper angulation) precluded reliable assessment of peri‐implant crestal bone levels.
2.4. Data Collection
Patient‐ and implant‐related information was manually extracted from the EHR database (AxiUm, Las Vegas, Nevada, USA) by two trained examiners (A.F. and N.V.) between May 2025 and September 2025. Data were obtained through a comprehensive review of treatment codes, clinical notes, and attached documents, including biomaterial specifications and radiographic records.
The following patient‐related variables were recorded: age at the time of implant placement (years); sex (male or female); ethnicity (Asian, Black or African American, White, or other); smoking status (current or former/never); presence of diabetes mellitus; osteoporosis; medication use (anti‐inflammatory, antidepressant, blood‐thinning, and/or antiresorptive medications [denosumab or bisphosphonates]); history of periodontitis (yes, no, or edentulous); and self‐reported penicillin allergy.
The following implant‐ and site‐related variables were collected: implant location (maxilla or mandible; posterior or anterior); operator (faculty or resident); implant manufacturer (Straumann, Astra, or other); implant diameter (narrow ≤ 3.6 mm, regular > 3.6 to < 4.8 mm, or wide ≥ 4.8 mm); implant length (short ≤ 8 mm or conventional > 8 mm); implant platform–abutment interface (bone‐level or tissue‐level); timing of implant placement (nonimmediate or immediate); healing mode (submerged/two‐stage or nonsubmerged/one‐stage); graft use at implant placement (yes or no); previous bone grafting (none/native bone, ridge preservation or guided bone regeneration [GBR]/sinus lift); graft material for bone augmentation (no graft, allograft, other); implant–tooth distance (inadequate [< 1.5 mm] or adequate [≥ 1.5 mm]) and implant–implant distance (inadequate < 3 mm or adequate ≥ 3 mm); insertion depth (equicrestal, subcrestal, or supracrestal); loading protocol (immediate or delayed); and antibiotic prescribed (none, penicillin/amoxicillin, clindamycin, azithromycin, or other [cephalexin and ciprofloxacin]).
2.5. Outcomes
Early implant failure was defined as implant loss occurring prior to functional loading as a result of failure to achieve osseointegration [31].
PLCBL was defined as a reduction of ≥ 0.5 mm in interproximal crestal bone level from implant placement to implant uncovering, abutment installation, or crown delivery. PLCBL was assessed on the mesial and distal aspects of each implant by a single trained and calibrated examiner (A.F.) using a calibrated linear measurement tool (MiPACS, Charlotte, North Carolina, USA) on digital radiographs (bitewing or periapical), as previously described [7]. To assess intra‐examiner reliability, a calibration exercise was performed in which crestal bone levels were measured on 10 radiographs at two separate time points, 7 days apart. The intraclass correlation coefficient was 0.93, indicating excellent agreement. PLCBL was categorized as ≥ 0.5 to < 1.5 mm or ≥ 1.5 mm.
Post‐implant placement surgical complications were defined as the presence of clinical signs of infection at the implant site, including pain not adequately controlled with analgesic medication, suppuration, and exacerbated local inflammation identified during postoperative follow‐up visits. These assessments typically occurred between 1 and 4 weeks after surgery and included visits for suture removal as well as additional visits prompted by patient‐reported pain or concerns related to the surgical site.
2.6. Statistical Analyzes
Descriptive statistics were used to summarize patient‐level and implant‐level characteristics and study outcomes. For continuous variables, means and standard deviations were reported. For categorical variables, frequencies and percentages were presented. Pearson's chi‐squared test was used for bivariate comparisons of early implant failure, PLCBL, and postoperative complications according to (1) self‐reported penicillin allergy and (2) prescription of penicillin. For pairwise comparisons across antibiotic types, Fisher's exact test was used, with Holm‐Bonferroni correction applied to adjust for multiple comparisons within each outcome.
Unadjusted mixed‐effects regression models were first fitted to estimate baseline associations prior to adjustment for potential confounders. Because individual patients could contribute more than one implant, all models included a random intercept for patient to account for within‐patient correlation. For implant failure and surgical complications, logistic mixed‐effects models were used, and results are presented as odds ratios (ORs), with OR > 1 indicating higher odds of the outcome in the exposed group. For PLCBL, categorized into ordered severity levels (no crestal bone loss, 0.5–1.5 mm, ≥ 1.5 mm), an ordinal mixed‐effects model was applied. In this model, the OR represents the odds of being in a more severe bone‐loss category across all thresholds. After evaluation of unadjusted associations, adjusted mixed‐effects regression models were fitted to account for potential confounding. A purposeful selection strategy was used to determine the final set of covariates for each exposure–outcome combination. For each outcome, a base model including the exposure of interest and prespecified clinically relevant covariates was specified. Additional candidate covariates were then evaluated individually by adding each variable to the base model. Covariates with p‐values < 0.25 and without convergence issues were retained for the multivariable stage. The final adjusted model included the exposure of interest, clinically relevant covariates, selected additional covariates, and a random intercept for patient to account for clustering of implants within individuals. For all models, unadjusted and adjusted ORs, 95% confidence intervals (CIs), and corresponding p‐values are reported.
This retrospective study included all eligible implants placed within the study period. Given the retrospective nature of this study, a sample size calculation was not performed, as we included all available data to maximize precision and representativeness.
All analyzes were performed using R (version 4.3.1). Mixed‐effects logistic regression models were fitted using the lme4 package, and ordinal mixed‐effects models were fitted using the ordinal package. Two‐sided p‐values < 0.05 were considered statistically significant.
3. Results
An initial database of 3336 dental charts was screened, and after removal of duplicate records, 2487 unique patient charts remained. From this dataset, approximately one‐fifth of the charts were randomly selected for detailed review. A total of 179 subjects and 374 implants were included in the final analysis. All implants were evaluated for early failure and surgical complications. Implants that failed were excluded from PLCBL assessment; consequently, 351 implants were available for PLCBL evaluation (Figure 1).
FIGURE 1.
Study flowchart.
Table S1 presents the distribution of patient‐related variables in the study population. The mean age at the time of implant placement was 62.3 ± 16.3 years. The majority of patients were female (53.6%), White (81.6%), former or never smokers (91.1%), and had a history of periodontitis (53.7%). Diabetes mellitus and osteoporosis were present in 12.3% and 3.9% of patients, respectively. The most commonly reported medication use was blood‐thinning agents (26.8%), followed by anti‐inflammatory medications (24.0%), antidepressants (16.2%), and antiresorptive medications (6.1%). One‐third of patients (33.0%) reported a penicillin allergy, which is consistent with the 1:2 sampling strategy used in this study.
Table S2 presents the distribution of implant‐related variables among the included implants. Most implants were placed in the maxilla (50.5%), in the posterior regions (74.6%), and were placed by residents (58.8%). The majority of implants were manufactured by Astra (57.8%) and Straumann (38.5%), had a regular diameter (65.8%), a conventional length (> 8 mm) (81.6%), and a bone‐level platform–abutment interface (81.8%). Most implants were placed in a non‐immediate manner (84.8%), healed using a one‐stage protocol (51.9%), placed in native bone (55.1%), and at a subcrestal depth (65.8%). Graft materials were used at the time of implant placement in 40.6% of cases, and 2.1% of implants were immediately loaded. Penicillin (primarily amoxicillin) was prescribed for 50.3% of implants, followed by clindamycin (16.0%), azithromycin (12.8%), and “other” antibiotics, including cephalexin and ciprofloxacin (6.4%); 14.5% of implant cases did not receive any antibiotic prescription.
Table S3 presents the prevalence of the study outcomes. PLCBL ≥ 0.5 mm was observed in 32.5% of implants, while early implant failure and post‐implant placement surgical complications occurred in 6.1% and 8.0% of implants, respectively.
Table 1 presents the bivariate comparisons of implant outcomes between patients with and without self‐reported penicillin allergy and according to penicillin prescription. A significantly higher proportion of implants with post‐implant placement surgical complications (15.5% vs. 3.1%) and early implant failure (12.8% vs. 1.8%) was observed among patients reporting a penicillin allergy compared with those without a reported allergy (p < 0.05). Furthermore, a significantly lower proportion of implants exhibited post‐implant placement surgical complications (3.2% vs. 12.9%) and early implant failure (1.6% vs. 10.8%) when penicillin was prescribed compared with when it was not prescribed (p < 0.05). No significant differences in PLCBL were observed either between patients with and without self‐reported penicillin allergy or according to penicillin prescription status (p > 0.05).
TABLE 1.
Implant outcome comparisons by self‐reported penicillin allergy and penicillin prescription.
| Outcomes | Self‐reported penicillin allergy | p a | |
|---|---|---|---|
| Yes (n = 148) | No (n = 226) | ||
| Surgical complication | 23 (15.5%) | 7 (3.1%) | < 0.001 |
| Preloading crestal bone loss | 0.366 | ||
| No EBL | 92 (71.3%) | 145 (65.3%) | |
| > 0.5 < 1.5 | 20 (15.5%) | 48 (21.6%) | |
| ≥ 1.5 | 17 (13.2%) | 29 (13.1%) | |
| Early implant failure | < 0.001 | ||
| No failure | 129 (87.2%) | 222 (98.2%) | |
| Early failure | 19 (12.8%) | 4 (1.8%) | |
| Penicillin prescription | |||
|---|---|---|---|
| Yes (n = 188) | No (n = 186) | ||
| Surgical complication | 6 (3.2%) | 24 (12.9%) | < 0.001 |
| Preloading crestal bone loss | 0.522 | ||
| No EBL | 122 (65.9%) | 115 (69.3%) | |
| > 0.5 < 1.5 | 40 (21.6%) | 28 (16.9%) | |
| ≥ 1.5 | 23 (12.4%) | 23 (13.9%) | |
| Early implant failure | < 0.001 | ||
| No failure | 85 (98.4%) | 166 (89.2%) | |
| Early failure | 3 (1.6%) | 20 (10.8%) | |
Note: Bold values indicate statistically significant results p < 0.05.
p‐values from Pearson's chi‐squared test.
Table 2 presents bivariate comparisons of implant outcomes across antibiotic prescription groups. After adjustment for multiple comparisons, a significantly higher frequency of surgical complications was observed in implants prescribed azithromycin compared with those prescribed penicillin (20.8% vs. 3.2%; adjusted p < 0.001). In contrast, the differences in surgical complication rates between penicillin and clindamycin and between azithromycin and clindamycin were no longer statistically significant after adjustment (p > 0.05). Regarding implant failure, implants prescribed azithromycin (20.8%) or clindamycin (8.3%) exhibited significantly higher rates of early implant failure compared with those prescribed penicillin (1.6%) after adjustment (p < 0.001 and p = 0.044, respectively). The difference in early implant failure rates between azithromycin and clindamycin was not statistically significant after adjustment (adjusted p = 0.092). No significant differences in PLCBL were observed among the different types of antibiotics prescribed (p > 0.05).
TABLE 2.
Implant outcome comparisons by antibiotic types.
| Outcomes | Penicillin versus azithromycin | Penicillin versus clindamycin | Azithromycin versus clindamycin | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Penicillin N = 188 a | Azithromycin N = 48 a | p b | Adjusted p c | Penicillin N = 188 a | Clindamycin N = 60 a | p b | Adjusted p c | Azithromycin N = 48 a | Clindamycin N = 60 a | p b | Adjusted p c | |
| Surgical complication | 6 (3.2%) | 10 (20.8%) | < 0.001 | < 0.001 | 6 (3.2%) | 6 (10.0%) | 0.043 | 0.086 | 10 (20.8%) | 6 (10.0%) | 0.172 | 0.172 |
| Preloading crestal bone loss | 0.523 | 0.704 | 0468 | |||||||||
| No EBL | 122 (65.9%) | 28 (73.7%) | 122 (65.9%) | 34 (61.8%) | 28 (73.7%) | 34 (61.8%) | ||||||
| > 0.5 < 1.5 | 40 (21.6%) | 5 (13.2%) | 40 (21.6%) | 12 (21.8%) | 5 (13.2%) | 12 (21.8%) | ||||||
| ≥ 1.5 | 23 (12.4%) | 5 (13.2%) | 23 (12.4%) | 9 (16.4%) | 5 (13.2%) | 9 (16.4%) | ||||||
| Early implant failure | < 0.001 | < 0.001 | 0.022 | 0.044 | 0.092 | 0.092 | ||||||
| No failure | 185 (98.4%) | 38 (79.2%) | 185 (98.4%) | 55 (91.7%) | 38 (79.2%) | 55 (91.7%) | ||||||
| Early failure | 3 (1.6%) | 10 (20.8%) | 3 (1.6%) | 5 (8.3%) | 10 (20.8%) | 5 (8.3%) | ||||||
Note: Bold values indicate statistically significant results p < 0.05.
Values are n (column %).
p‐values from Fisher's exact test (unadjusted).
Holm–Bonferroni adjusted p‐values for three pairwise comparisons of antibiotic type for each outcome.
According to the mixed‐effects logistic regression models, neither the unadjusted nor the adjusted analyzes demonstrated statistically significant associations between self‐reported penicillin allergy and surgical complications (p > 0.05; Table 3), PLCBL (p > 0.05; Table 4), or early implant failure (p > 0.05; Table 5). Furthermore, none of the covariates included in the models showed statistically significant associations with surgical complications (p > 0.05; Table 3) or early implant failure (p > 0.05; Table 5). In contrast, tissue‐level implants were associated with significantly lower odds of PLCBL compared with bone‐level implants (OR = 0.23; 95% CI: 0.09–0.63; p = 0.004; Table 4).
TABLE 3.
Mixed‐effects logistic regression analysis of the association between self‐reported penicillin allergy and surgical complications.
| Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|
| OR a | 95% CI b | p | OR a | 95% CI b | p | |
| Self‐reported penicillin allergy | ||||||
| No | — | — | — | — | ||
| Yes | 5.94 | 0.38, 93.2 | 0.205 | 25.0 | 0.13, 4884 | 0.232 |
| Smoking status | ||||||
| Current | — | — | ||||
| Former/never | 0.60 | 0.00, 971 | 0.892 | |||
| Diabetes | ||||||
| Yes | — | — | ||||
| No | 0.51 | 0.00, 155 | 0.816 | |||
| History of periodontitis | ||||||
| No | — | — | ||||
| Yes | 1.00 | 0.01, 99.9 | > 0.999 | |||
| Moment of implant placement | ||||||
| Non‐immediate | — | — | ||||
| Immediate | 3.70 | 0.19, 100 | 0.389 | |||
| Healing mode | ||||||
| Submerged | — | — | ||||
| Nonsubmerged | 23.1 | 0.59, 899 | 0.093 | |||
| Implant depth | ||||||
| Equicrestal | — | — | ||||
| Subcrestal | 27.0 | 0.23, 3225 | 0.177 | |||
| Supracrestal | 0.01 | 0.00, 60.8 | 0.312 | |||
| Implant brand | ||||||
| Straumann | — | — | ||||
| Astra | 4.61 | 0.04, 605 | 0.539 | |||
| Other | 2407 | 0.60, 9 681 783 | 0.066 | |||
OR = Odds Ratio from mixed‐effects model with random intercept for patient.
CI = Confidence interval.
TABLE 4.
Mixed‐effects logistic regression analysis of the association between self‐reported penicillin allergy and PLCBL.
| Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|
| OR a | 95% CI b | p | OR a | 95% CI b | p | |
| Self‐reported penicillin allergy | ||||||
| No | — | — | — | — | ||
| Yes | 0.78 | 0.36, 1.69 | 0.530 | 0.84 | 0.38, 1.84 | 0.668 |
| Smoking status | ||||||
| Current | — | — | ||||
| Former/never | 1.49 | 0.43, 5.14 | 0.525 | |||
| Diabetes | ||||||
| Yes | — | — | ||||
| No | 1.55 | 0.51, 4.75 | 0.439 | |||
| History of periodontitis | ||||||
| No | — | — | ||||
| Yes | 0.98 | 0.46, 2.11 | 0.966 | |||
| Implant level | ||||||
| Bone level | — | — | ||||
| Tissue level | 0.23 | 0.09, 0.63 | 0.004 | |||
| Moment of implant placement | ||||||
| Nonimmediate | — | — | ||||
| Immediate | 1.25 | 0.50, 3.13 | 0.629 | |||
| Previous bone augmentation | ||||||
| No | — | — | ||||
| Ridge preservation | 0.55 | 0.25, 1.21 | 0.135 | |||
| GBR and/or sinus | 0.63 | 0.22, 1.79 | 0.389 | |||
Note: Bold values indicate statistically significant results p < 0.05.
OR = Odds ratio from mixed‐effects model with random intercept for patient.
CI = Confidence interval.
TABLE 5.
Mixed‐effects logistic regression analysis of the association between self‐reported penicillin allergy and early implant failure.
| Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|
| OR a | 95% CI b | p | OR a | 95% CI b | p | |
| Self‐reported penicillin allergy | ||||||
| No | — | — | — | — | ||
| Yes | 5.99 | 0.22, 161 | 0.286 | 12.3 | 0.35, 427 | 0.166 |
| Diabetes | ||||||
| Yes | — | — | ||||
| No | 3.59 | 0.02, 785 | 0.642 | |||
| History of periodontitis | ||||||
| No | — | — | ||||
| Yes | 0.42 | 0.02, 11.3 | 0.608 | |||
| Moment of implant placement | ||||||
| Nonimmediate | — | — | ||||
| Immediate | 9.09 | 0.75, 100 | 0.083 | |||
| Previous bone augmentation | ||||||
| No | — | — | ||||
| Ridge preservation | 0.54 | 0.03, 10.8 | 0.690 | |||
| GBR and/or sinus | 0.93 | 0.04, 21.9 | 0.965 | |||
OR = Odds ratio from mixed‐effects model with random intercept for patient.
CI = Confidence interval.
In both unadjusted and adjusted analyzes, azithromycin (unadjusted OR = 40.3, 95% CI: 2.08–778; adjusted OR = 192, 95% CI: 2.82–13 116; p < 0.05) and “other” antibiotics (unadjusted OR = 211, 95% CI: 6.26–7104; adjusted OR = 345, 95% CI: 2.80–42 560; p < 0.05) were associated with significantly higher odds of post‐implant placement surgical complications compared with penicillin. Clindamycin was not significantly associated with surgical complications in either analysis (p > 0.05) (Table 6).
TABLE 6.
Mixed‐effects logistic regression analysis of the association between type of antibiotic prescribed and surgical complications.
| Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|
| OR a | 95% CI b | p | OR a | 95% CI b | p | |
| Antibiotic types | ||||||
| Penicillin | — | — | — | — | ||
| Azithromycin | 40.3 | 2.08, 778 | 0.014 | 192 | 2.82, 13 116 | 0.015 |
| Clindamycin | 2.07 | 0.10, 43.4 | 0.640 | 3.66 | 0.08, 178 | 0.513 |
| Other | 211 | 6.26, 7104 | 0.003 | 345 | 2.80, 42 560 | 0.017 |
| Smoking status | ||||||
| Current | — | — | ||||
| Former/never | 0.36 | 0.00, 202 | 0.754 | |||
| Diabetes | ||||||
| Yes | — | — | ||||
| No | 0.49 | 0.01, 33.4 | 0.741 | |||
| History of periodontitis | ||||||
| No | — | — | ||||
| Yes | 0.51 | 0.02, 12.9 | 0.686 | |||
| Moment of implant placement | ||||||
| Nonimmediate | — | — | ||||
| Immediate | 8.33 | 0.72, 100 | 0.090 | |||
| Healing mode | ||||||
| Submerged | — | — | ||||
| Nonsubmerged | 6.73 | 0.71, 64.0 | 0.097 | |||
| Implant brand | ||||||
| Straumann | — | — | ||||
| Astra | 6.14 | 0.12, 312 | 0.366 | |||
| Other | 129 | 0.58, 28 678 | 0.078 | |||
Note: Bold values indicate statistically significant results p < 0.05.
OR = Odds ratio from mixed‐effects model with random intercept for patient; the group that did not take any antibiotics was excluded from complications model due to quasi‐separation.
CI = Confidence interval.
In the mixed‐effects logistic regression analyzes, no significant associations were observed between the type of antibiotic prescribed and PLCBL (Table 7) and implant failure (Table 8), using penicillin as the reference category, in either unadjusted or adjusted models (p > 0.05 for all comparisons). Tissue‐level implants were associated with significantly lower odds of PLCBL compared with bone‐level implants (OR = 0.22; 95% CI: 0.08–0.62; p = 0.004) (Table 7). None of the patient‐ or implant‐related covariates included in the model were significantly associated with early implant failure (p > 0.05; Table 8).
TABLE 7.
Mixed‐effects logistic regression analysis of the association between type of antibiotic prescribed and PLCBL.
| Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|
| OR a | 95% CI b | p | OR a | 95% CI b | p | |
| Antibiotic types | ||||||
| Penicillin | — | — | — | — | ||
| No antibiotic | 1.11 | 0.43, 2.90 | 0.824 | 1.19 | 0.45, 3.14 | 0.729 |
| Azithromycin | 0.77 | 0.23, 2.57 | 0.670 | 1.00 | 0.29, 3.48 | > 0.999 |
| Clindamycin | 1.36 | 0.49, 3.77 | 0.552 | 1.58 | 0.56, 4.47 | 0.389 |
| Other | 0.91 | 0.16, 5.24 | 0.917 | 0.73 | 0.13, 4.14 | 0.725 |
| Smoking status | ||||||
| Current | — | — | ||||
| Former/never | 1.37 | 0.40, 4.73 | 0.619 | |||
| Diabetes | ||||||
| Yes | — | — | ||||
| No | 1.48 | 0.47, 4.61 | 0.502 | |||
| History of periodontitis | ||||||
| No | — | — | ||||
| Yes | 0.96 | 0.45, 2.09 | 0.927 | |||
| Implant level | ||||||
| Bone level | — | — | ||||
| Tissue level | 0.22 | 0.08, 0.62 | 0.004 | |||
| Moment of implant placement | ||||||
| Nonimmediate | — | — | ||||
| Immediate | 1.32 | 0.53, 3.33 | 0.553 | |||
| Previous bone augmentation | ||||||
| No | — | — | ||||
| Ridge preservation | 0.56 | 0.25, 1.23 | 0.147 | |||
| GBR and/or sinus | 0.64 | 0.22, 1.89 | 0.422 | |||
Note: Bold values indicate statistically significant results p < 0.05.
OR = Odds ratio from mixed‐effects model with random intercept for patient.
CI = Confidence interval.
TABLE 8.
Mixed‐effects logistic regression analysis of the association between type of antibiotic prescribed and early implant failure.
| Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|
| OR a | 95% CI b | p | OR a | 95% CI b | p | |
| Antibiotic types | ||||||
| Penicillin | — | — | — | — | ||
| No antibiotic | 1.34 | 0.05, 38.3 | 0.864 | 0.88 | 0.03, 30.9 | 0.944 |
| Azithromycin | 7.32 | 0.19, 284 | 0.286 | 46.4 | 0.64, 3364 | 0.079 |
| Clindamycin | 0.99 | 0.01, 87.6 | 0.998 | 1.09 | 0.01, 143 | 0.973 |
| Other | 7.78 | 0.16, 374 | 0.299 | 1.98 | 0.01, 342 | 0.795 |
| Diabetes | ||||||
| Yes | — | — | ||||
| No | 1.74 | 0.01, 503 | 0.849 | |||
| History of periodontitis | ||||||
| No | — | — | ||||
| Yes | 0.35 | 0.01, 11.3 | 0.557 | |||
| Moment of implant placement | ||||||
| Nonimmediate | — | — | ||||
| Immediate | 10.0 | 0.75, 100 | 0.080 | |||
| Previous bone augmentation | ||||||
| No | — | — | ||||
| Ridge preservation | 0.54 | 0.02, 12.0 | 0.698 | |||
| GBR and/or sinus | 0.08 | 0.00, 8.36 | 0.289 | |||
OR = Odds ratio from mixed‐effects model with random intercept for patient.
CI = Confidence interval.
Tables S4–S6 present the mixed‐effects logistic regression models evaluating the association between penicillin prescription and surgical complications, PLCBL, and implant failure, respectively. Overall, neither the unadjusted nor the adjusted analyzes provided evidence of a significant association between penicillin prescription and surgical complications, PLCBL, or implant failure (p > 0.05). Tissue‐level implants were associated with lower odds of PLCBL compared with bone‐level implants (OR = 0.22; 95% CI: 0.08–0.62; p = 0.004).
4. Discussion
This retrospective cohort study investigated the association between self‐reported penicillin allergy and the use of alternative antibiotics with early dental implant outcomes. The results indicate that self‐reported penicillin allergy was associated with a higher frequency of surgical complications and early implant failure in unadjusted analyzes; however, this association was no longer observed after adjustment for relevant confounders. Among the antibiotics assessed, the prescription of azithromycin and “other” alternative antibiotics, including cephalexin and ciprofloxacin, remained independently associated with surgical complications. These findings suggest that the prescription of alternative antibiotic regimens, rather than self‐reported penicillin allergy itself, may be associated with early implant outcomes.
Self‐reported penicillin allergy, often not reflecting true immunologic hypersensitivity, is clinically relevant because it frequently leads to the use of less effective alternative antibiotics. In the present study, higher frequencies of surgical complications and implant failure were observed among cases with self‐reported penicillin allergy and among those not prescribed penicillin. These findings align with prior studies showing higher implant failure and complication rates in patients with self‐reported penicillin allergy and lower rates with amoxicillin prescriptions, even when given as a single preoperative dose [26, 27, 30, 32, 33]. However, after adjustment for potential confounders, neither self‐reported penicillin allergy nor penicillin prescription independently predicted postoperative complications or implant failure. Instead, self‐reported penicillin allergy appears to act as a surrogate for a broader clinical context, including patient characteristics, comorbidities, case complexity, and treatment‐related factors, particularly the use of alternative antibiotic regimens, rather than as an independent causal risk factor. Studies employing regression analyzes to account for confounding factors have yielded inconsistent findings regarding the association between self‐reported penicillin allergy and implant complications. Consistent with our findings, a previous study observed an association between penicillin allergy and implant failure in univariate analyzes, which was not sustained in multivariable models [25]. In contrast, a subsequent study with a larger sample size reported a significant association even after adjustment for confounders [26]. These differences likely reflect methodological variations across studies, including definitions of failure (early, late, or combined), sample size, regression approach (mixed‐ vs. fixed‐effects), and the variables included in the models. Overall, the heterogeneity of existing evidence underscores the need for further studies to clarify the relationships between self‐reported penicillin allergy, penicillin use, and implant outcomes.
Beyond evaluating self‐reported penicillin allergy and penicillin prescription, this study also examined the specific alternative antibiotics prescribed. Clindamycin and azithromycin were the most frequently prescribed antibiotics among patients reporting penicillin allergy. Higher frequencies of complications and early failure were observed among implants placed in patients prescribed azithromycin or clindamycin compared with penicillin. In exploratory multivariable analyzes adjusting for potential confounders, azithromycin and “other” alternative antibiotics, including cephalexin and ciprofloxacin, remained associated with higher odds of surgical complications compared with penicillin (Table 6). Furthermore, azithromycin showed a trend toward increased odds of early implant failure after adjustment (Table 8). Given the limited sample size of some antibiotic subgroups, these findings should be interpreted cautiously. The association between azithromycin use and implant‐related outcomes remains poorly explored in the literature. A previous study suggested that azithromycin may improve early implant healing compared with amoxicillin; however, it was limited by a small sample size, short follow‐up (20 days), and lack of assessment of implant failure or surgical complications [34]. More recently, Chatzopoulos and Wolff [35] reported no differences in failure rates between preoperative azithromycin and amoxicillin in immediate dental implants. Notably, azithromycin was prescribed primarily using the standard 5‐day regimen (500 mg on Day 1, followed by 250 mg daily on Days 2–5); therefore, the findings related to azithromycin reflect this specific dosing protocol, and alternative regimens could not be evaluated. Consistent with the present findings, previous investigations have reported higher implant failure rates among penicillin‐allergic patients treated with clindamycin compared with those treated with amoxicillin [12, 27, 30, 35], as well as an increased risk of postoperative infection [12]. Interestingly, Chatzopoulos and Wolff [35] recently demonstrated that ciprofloxacin, which comprised a large part of the “other” antibiotics group in our study, was associated with higher failure rates in immediate dental implants. Although these findings are clinically relevant and hypothesis‐generating, the evidence remains preliminary, and comparisons across studies are limited by differences in clinical contexts (immediate vs. nonimmediate implants) and antibiotic protocols (timing, dosage, and duration). This reinforces the need for well‐designed prospective controlled clinical studies to better elucidate the impact of alternative antibiotic regimens on implant‐related outcomes.
In our previous study [7], self‐reported penicillin allergy was associated with increased rates of PLCBL and remained significant for PLCBL ≥ 1.5 mm in regression models, leading us to propose that further investigations were warranted to validate these findings. Thus, the present study was also designed to re‐examine this association in an independent cohort. However, the relationship between self‐reported penicillin allergy and PLCBL was not reproduced. This discrepancy may reflect the smaller number of allergy‐reported patients and implants in the earlier cohort, potentially leading to overestimation of effects, as well as the use of more robust statistical models in the current study that account for within‐patient correlation. These findings underscore the need for replication and support cautious interpretation of self‐reported penicillin allergy as a risk factor for PLCBL.
A major strength of this study is its comprehensive assessment of self‐reported penicillin allergy in relation to three key early implant outcomes, along with the effects of alternative antibiotic regimens. In addition, the use of multivariable regression modeling to adjust for multiple patient‐ and procedural‐related confounders represents a significant methodological strength. This approach demonstrated that several associations identified in bivariate analyzes did not persist after adjustment, underscoring the importance of controlling for confounding. Notably, tissue‐level implants consistently demonstrated significantly lower odds of PLCBL compared with bone‐level implants. This finding supports the clinical benefit of positioning the implant–abutment connection and its associated microgap at a distance from the crestal bone, as previously reported [7, 35, 36, 37].
This study has limitations inherent to its retrospective design. First, antibiotic adherence could not be verified; therefore, patient compliance with the prescribed regimens documented in the dental records could not be confirmed. In addition, the relatively small number of cases within antibiotic subgroups resulted in wide confidence intervals, limiting the precision of these estimates. Therefore, associations involving specific alternative antibiotics, particularly azithromycin and “other” antibiotics, should be interpreted with caution and considered hypothesis‐generating. Furthermore, penicillin allergy and other medical conditions were self‐reported. While this reflects real‐world clinical practice, where antibiotic prescribing decisions are frequently based on patient‐reported history rather than confirmatory testing, it limits the ability to determine whether the observed associations are attributable to true allergy status or to differences in antibiotic prescribing patterns. In this context, the present findings likely reflect the nonprescription of penicillin following a reported allergy, rather than the biological effect of allergy itself. Finally, the single‐center, university‐based setting may limit the generalizability of the findings to other clinical environments with different patient profiles and prescribing practices.
The biological mechanisms underlying our findings warrant further investigation. One possible explanation is that self‐reported penicillin allergy may be associated with other hypersensitivity conditions, including titanium allergy; however, this study could not determine which patients had a true immunologically confirmed penicillin allergy. The observed associations with alternative antibiotics may reflect reduced antimicrobial efficacy, potential effects on wound healing (e.g., impaired osteoblast function reported for clindamycin and azithromycin) [38, 39], or differences in dosing and duration. Based on our findings, early implant morbidity appears more closely related to the choice of alternative antibiotics and the clinical context in which they are used, particularly in medically complex patients and more demanding scenarios (e.g., immediate placement, augmented sites, limited bone volume), than to self‐reported penicillin allergy itself. Clinically, these findings highlight the potential relevance of antibiotic prescribing patterns in patients reporting penicillin allergy undergoing implant placement. In this context, penicillin allergy verification (delabeling) may help reduce unnecessary use of alternative antibiotics. From a research perspective, multicenter prospective studies and, ideally, randomized controlled trials with standardized antibiotic protocols, adherence verification, larger subgroups, and validated penicillin allergy assessments are needed. These studies should clarify the independent effects of reported allergy, true allergy, and specific alternative antibiotics regimens on implant outcomes.
5. Conclusion
Self‐reported penicillin allergy does not appear to be an independent determinant of early implant outcomes but may instead reflect the broader clinical context in which alternative antibiotic regimens are prescribed. In contrast, the use of alternative antibiotics may be associated with early postoperative outcomes, although these observations should be interpreted with caution given the limited sample size within subgroups. Further studies are needed to clarify the potential association between alternative antibiotic use and early implant complications.
Author Contributions
Adrienne Fetner: data collection, critical revision of article, approval of article. Nathalia Vilela: data collection, drafting article, critical revision of article, approval of article. Xinlin Lu: statistics, data analysis/interpretation, critical revision of article, approval of article. Guogen Shan: statistics, data analysis/interpretation, critical revision of article, approval of article. Poliana M. Duarte: concept/design, data analysis/interpretation, drafting article, critical revision of article, approval of article.
Funding
The authors have nothing to report.
Ethics Statement
The study was approved by the University of Florida Institutional Review Board (IRB#202101382) and conducted in accordance with the observational studies in epidemiology (STROBE) guidelines. This study was approved as exempt because it poses minimal risk and was approved under the following exempt category/categories: (4)(iii) Secondary research for which consent is not required.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Data S1: STROBE statement—checklist of items that should be included in reports of cohort studies.
Table S1: Patient‐related variables.
Table S2: Implant/surgery‐related variables.
Table S3: Implant‐related outcomes.
Table S4: Mixed‐effects logistic regression analysis of the association between penicillin prescription and surgical complications.
Table S5: Mixed‐effects logistic regression analysis of the association between penicillin prescription and PLCBL.
Table S6: Mixed‐effects logistic regression analysis of the association between penicillin prescription and early implant failure.
Acknowledgments
The authors have nothing to report.
Fetner A., Vilela N., Lu X., Shan G., and Duarte P. M., “Self‐Reported Penicillin Allergy, Alternative Antibiotics, and Implant Outcomes: A Retrospective Study,” Clinical Implant Dentistry and Related Research 28, no. 4 (2026): e70169, 10.1111/cid.70169.
Adrienne Fetner and Nathalia Vilela contributed to the manuscript equally.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Schimmel M., Araujo M., Abou‐Ayash S., et al., “Group 4 ITI Consensus Report: Patient Benefits Following Implant Treatment in Partially and Fully Edentulous Patients,” Clinical Oral Implants Research 34, no. S26 (2023): 257–265, 10.1111/clr.14145. [DOI] [PubMed] [Google Scholar]
- 2. French D., Ofec R., and Levin L., “Long Term Clinical Performance of 10 871 Dental Implants With up to 22 Years of Follow‐Up: A Cohort Study in 4247 Patients,” Clinical Implant Dentistry and Related Research 23, no. 3 (2021): 289–297, 10.1111/cid.12994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Lupi S. M., Agolli J., Isola G., and Scribante A., “Dental Implant Survival and Marginal Bone Loss After a Minimum of 20 Years: Systematic Review and Meta‐Analysis of Prospective Studies,” International Journal of Oral and Maxillofacial Surgery (2026), Online ahead of print, 10.1016/j.ijom.2026.05.011. [DOI] [PubMed] [Google Scholar]
- 4. Kupka J. R., König J., Al‐Nawas B., Sagheb K., and Schiegnitz E., “How Far Can We Go? A 20‐Year Meta‐Analysis of Dental Implant Survival Rates,” Clinical Oral Investigations 28, no. 10 (2024): 541, 10.1007/s00784-024-05929-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Dionigi C., Bougas K., Berglundh T., and Derks J., “Marginal Bone Loss Around Implants – What Does It Mean?,” Periodontology 2000 (2026): 1–41, 10.1111/prd.70036. [DOI] [PubMed] [Google Scholar]
- 6. Lin G., Ye S., Liu F., and He F., “A Retrospective Study of 30,959 Implants: Risk Factors Associated With Early and Late Implant Loss,” Journal of Clinical Periodontology 45, no. 6 (2018): 733–743, 10.1111/jcpe.12898. [DOI] [PubMed] [Google Scholar]
- 7. Vilela N., Gurgel B. C. V., de Bruzos C., et al., “Preloading Peri‐Implant Crestal Bone Loss: A Retrospective Study of Incidence and Related Factors,” Journal of Periodontology 95, no. 10 (2024): 963–976, 10.1002/JPER.24-0028. [DOI] [PubMed] [Google Scholar]
- 8. Guarnieri R., Reda R., di Nardo D., Miccoli G., Zanza A., and Testarelli L., “Analysis of Risk Factors Related to Early Implant Failures in Patients Attending a Private Practice Setting: A Retrospective Study,” Journal of Clinical Medicine 14, no. 18 (2025): 6546, 10.3390/jcm14186546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Deeb G. R., Soung G. Y., Best A. M., and Laskin D. M., “Antibiotic Prescribing Habits of Oral and Maxillofacial Surgeons in Conjunction With Routine Dental Implant Placement,” Journal of Oral and Maxillofacial Surgery 73, no. 10 (2015): 1926–1931, 10.1016/j.joms.2015.05.024. [DOI] [PubMed] [Google Scholar]
- 10. Choi Y. Y., “Prescription of Antibiotics After Tooth Extraction in Adults: A Nationwide Study in Korea,” Journal of the Korean Association of Oral and Maxillofacial Surgeons 46, no. 1 (2020): 49–57, 10.5125/jkaoms.2020.46.1.49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Hai J. H., Lee C., Kapila Y. L., Chaffee B. W., and Armitage G. C., “Antibiotic Prescribing Practices in Periodontal Surgeries With and Without Bone Grafting,” Journal of Periodontology 91, no. 4 (2020): 508–515, 10.1002/JPER.19-0195. [DOI] [PubMed] [Google Scholar]
- 12. Salgado‐Peralvo A. O., Peña‐Cardelles J. F., Kewalramani N., et al., “Is Penicillin Allergy a Risk Factor for Early Dental Implant Failure? A Systematic Review,” Antibiotics 10, no. 10 (2021): 1227, 10.3390/antibiotics10101227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Sbricoli L., Grisolia G., Stellini E., Bacci C., Annunziata M., and Bressan E., “Antibiotic‐Prescribing Habits in Dentistry: A Questionnaire‐Based Study,” Antibiotics 13, no. 2 (2024): 189, 10.3390/antibiotics13020189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Golchini P., Behkar S., Kocamaz Ö. F., and Altundoğan S., “A Cross‐Sectional Survey of Dentists on Antibiotic Prescribing and Allergy Management in Oral Surgery,” Oral Health & Preventive Dentistry 23 (2025): 805–813, 10.3290/j.ohpd.c_2391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Blumenthal K. G., Ryan E. E., Li Y., Lee H., Kuhlen J. L., and Shenoy E. S., “The Impact of a Reported Penicillin Allergy on Surgical Site Infection Risk,” Clinical Infectious Diseases 66, no. 3 (2018): 329–336, 10.1093/cid/cix794. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Shenoy E. S., Macy E., Rowe T., and Blumenthal K. G., “Evaluation and Management of Penicillin Allergy,” Journal of the American Medical Association 321, no. 2 (2019): 188–199, 10.1001/jama.2018.19283. [DOI] [PubMed] [Google Scholar]
- 17. Antoon J. W., Grijalva C. G., Grisso A. G., et al., “Feasibility of a Centralized, Pharmacy‐Led Penicillin Allergy Delabeling Program,” Hospital Pediatrics 12, no. 7 (2022): e230–e237, 10.1542/hpeds.2021-006369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Lozo S., Wagner D., Shah N., Goldberg R., Gafni‐Kane A., and Solomonides A., “Should Penicillin Allergy Testing be Included as Part of Preoperative Testing?,” Journal for Healthcare Quality 45, no. 5 (2023): 255–260, 10.1097/JHQ.0000000000000395. [DOI] [PubMed] [Google Scholar]
- 19. Aloyan T., Salimova D., Antonov E., Chandra G., and McGrath K., “Penicillin Allergy Delabeling: Provider Knowledge, Practice Patterns, and Barriers Among Internal Medicine Residents in a Large Urban Community Teaching Hospital,” Cureus 17, no. 11 (2025): e97928, 10.7759/cureus.97928. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Centers for Disease Control and Prevention , “Clinical Features of Penicillin Allergy,” accessed December 20, 2025, https://www.cdc.gov/antibiotic‐use/hcp/clinical‐signs/index.html.
- 21. Luintel A., Healy J., Blank M., et al., “The Global Prevalence of Reported Penicillin Allergy: A Systematic Review and Meta‐Analysis,” Journal of Infection 90, no. 2 (2025): 106429, 10.1016/j.jinf.2025.106429. [DOI] [PubMed] [Google Scholar]
- 22. Khoury F., Javed F., and Romanos G., “Sinus Augmentation Failure and Postoperative Infections Associated With Prophylactic Clindamycin Therapy: An Observational Case Series,” International Journal of Oral & Maxillofacial Implants 33, no. 5 (2018): 1136–1139, 10.11607/jomi.6517. [DOI] [PubMed] [Google Scholar]
- 23. Basma H. and Misch C., “Extraction Socket Grafting and Ridge Augmentation Failures Associated With Clindamycin Antibiotic Therapy: A Retrospective Study,” International Journal of Oral and Maxillofacial Implants 36, no. 1 (2021): 122–125, 10.11607/jomi.8461. [DOI] [PubMed] [Google Scholar]
- 24. Roistacher D. M., Heller J. A., Ferraro N. F., and August M., “Is Penicillin Allergy a Risk Factor for Surgical Site Infection After Oral and Maxillofacial Surgery?,” Journal of Oral and Maxillofacial Surgery 80, no. 1 (2022): 93–100, 10.1016/j.joms.2021.08.147. [DOI] [PubMed] [Google Scholar]
- 25. French D., Larjava H., and Ofec R., “Retrospective Cohort Study of 4591 Straumann Implants in Private Practice Setting, With up to 10‐Year Follow‐Up. Part 1: Multivariate Survival Analysis,” Clinical Oral Implants Research 26, no. 11 (2015): 1345–1354, 10.1111/clr.12463. [DOI] [PubMed] [Google Scholar]
- 26. French D., Noroozi M., Shariati B., and Larjava H., “Clinical Retrospective Study of Self‐Reported Penicillin Allergy on Dental Implant Failures and Infections,” Quintessence International 47, no. 10 (2016): 861–870, 10.3290/j.qi.a36887. [DOI] [PubMed] [Google Scholar]
- 27. Salomó‐Coll O., Lozano‐Carrascal N., Lázaro‐Abdulkarim A., Hernández‐Alfaro F., Gargallo‐Albiol J., and Satorres‐Nieto M., “Do Penicillin‐Allergic Patients Present a Higher Rate of Implant Failure?,” International Journal of Oral & Maxillofacial Implants 33, no. 6 (2018): 1390–1395, 10.11607/jomi.7018. [DOI] [PubMed] [Google Scholar]
- 28. Block M. S., Christensen B. J., Mercante D. E., and Chapple A. G., “What Factors Are Associated With Implant Failure?,” Journal of Oral and Maxillofacial Surgery 79, no. 1 (2021): 91–97, 10.1016/j.joms.2020.08.023. [DOI] [PubMed] [Google Scholar]
- 29. Edibam N. R., Lorenzo‐Pouso A. I., and Caponio V. C. A., “Self‐Reported Allergy to Penicillin and Clindamycin Administration May Be Risk Factors for Dental Implant Failure: A Systematic Review, Meta‐Analysis and Delabeling Protocol,” Clinical Oral Implants Research 34, no. 7 (2023): 651–661, 10.1111/clr.14073. [DOI] [PubMed] [Google Scholar]
- 30. Zahra B., Nicholas B., Geoffrey R., Dina Z., Janal M. N., and Stuart F., “Dental Implant Failure Rates in Patients With Self‐Reported Allergy to Penicillin,” Clinical Implant Dentistry and Related Research 24, no. 3 (2022): 301–306, 10.1111/cid.13082. [DOI] [PubMed] [Google Scholar]
- 31. Esposito M., Hirsch J., Lekholm U., and Thomsen P., “Biological Factors Contributing to Failures of Osseointegrated Oral Implants, (I). Success Criteria and Epidemiology,” European Journal of Oral Sciences 106, no. 1 (1998): 527–551, 10.1046/j.0909-8836..t01-2-.x. [DOI] [PubMed] [Google Scholar]
- 32. Nolan R., Kemmoona M., Polyzois I., and Claffey N., “The Influence of Prophylactic Antibiotic Administration on Post‐Operative Morbidity in Dental Implant Surgery. A Prospective Double Blind Randomized Controlled Clinical Trial,” Clinical Oral Implants Research 25, no. 2 (2014): 252–259, 10.1111/clr.12124. [DOI] [PubMed] [Google Scholar]
- 33. Bravo‐Olmedo F., Reyes‐Botella C., Ocaña‐Peinado F. M., Manzano‐Moreno F. J., de Romero‐Olid M. N., and Olmedo‐Gaya M. V., “Effect of Antibiotic Prophylaxis in Dental Implant Surgery: A Randomized Controlled Clinical Trial,” Dentistry Journal 13, no. 11 (2025): 500, 10.3390/dj13110500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Escalante M. G., Eubank T. D., Leblebicioglu B., and Walters J. D., “Comparison of Azithromycin and Amoxicillin Before Dental Implant Placement: An Exploratory Study of Bioavailability and Resolution of Postoperative Inflammation,” Journal of Periodontology 86, no. 11 (2015): 1190–1200, 10.1902/jop.2015.150024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Chatzopoulos G. S. and Wolff L. F., “Comparing the Impact of Pre‐Operative Antibiotics on the Outcomes of Immediately Placed Dental Implants: A Retrospective Multi‐Center Study,” Methods and Protocols 8, no. 4 (2025): 69, 10.3390/mps8040069. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. van Eekeren P., Tahmaseb A., and Wismeijer D., “Crestal Bone Changes in Macrogeometrically Similar Implants With the Implant–Abutment Connection at the Crestal Bone Level or 2.5 Mm Above: A Prospective Randomized Clinical Trial,” Clinical Oral Implants Research 27, no. 12 (2016): 1479–1484, 10.1111/clr.12581. [DOI] [PubMed] [Google Scholar]
- 37. Agustín‐Panadero R., Bermúdez‐Mulet I., Fernández‐Estevan L., et al., “Peri‐Implant Behavior of Tissue Level Dental Implants With a Convergent Neck,” International Journal of Environmental Research and Public Health 18, no. 10 (2021): 5232, 10.3390/ijerph18105232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Naal F. D., Salzmann G. M., von Knoch F., et al., “The Effects of Clindamycin on Human Osteoblasts in Vitro,” Archives of Orthopaedic and Trauma Surgery 128, no. 3 (2008): 317–323, 10.1007/s00402-007-0561-y. [DOI] [PubMed] [Google Scholar]
- 39. Kato K., Ozaki M., Nakai K., et al., “Effect of Azithromycin on Mineralized Nodule Formation in MC3T3‐E1 Cells,” Current Issues in Molecular Biology 43, no. 3 (2021): 1451–1459, 10.3390/cimb43030102. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1: STROBE statement—checklist of items that should be included in reports of cohort studies.
Table S1: Patient‐related variables.
Table S2: Implant/surgery‐related variables.
Table S3: Implant‐related outcomes.
Table S4: Mixed‐effects logistic regression analysis of the association between penicillin prescription and surgical complications.
Table S5: Mixed‐effects logistic regression analysis of the association between penicillin prescription and PLCBL.
Table S6: Mixed‐effects logistic regression analysis of the association between penicillin prescription and early implant failure.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.