Influence of Finishing Clear Aligner Marginal Termination Design on Subgingival Periodontal Microbiota During Orthodontic Treatment: A Split-Mouth Study.

\- BACKGROUND AND RATIONALE Clear aligner therapy is widely used in orthodontics. Two trimming-line designs are commonly employed during the finishing phase. The supragingival design uses a straight horizontal cut positioned approximately 2 mm apical to the free gingival margin: the aligner edge covers approximately 2 mm of gingival tissue with direct contact but does not penetrate the gingival sulcus. The juxtagingival festooned design uses a scalloped cut that follows the natural contour of the free gingival margin exactly, terminating at the gingival margin without covering gingival tissue and without entering the sulcus. The two designs differ fundamentally in their relationship with the gingival margin. The supragingival design creates a semi-enclosed microenvironment at the sulcus entrance: by covering 2 mm of gingival tissue, the aligner limits access of saliva and atmospheric oxygen to the sulcus entrance region. Reduced salivary clearance and decreased oxygen tension are established conditions that favor anaerobic bacterial growth in the periodontal environment. Whether 2 mm of gingival tissue coverage by the aligner edge is sufficient to produce a measurable change in sulcular oxygen tension or salivary clearance has not been directly demonstrated and constitutes the biological hypothesis under investigation. The juxtagingival design, following the gingival contour at its margin, leaves the sulcus entrance more accessible to saliva and oxygen, potentially maintaining less favorable conditions for anaerobic colonization. These microenvironmental differences have not been investigated in published studies. No prior study has examined whether trimming-line geometry affects subgingival periodontal microbiota composition in patients undergoing aligner therapy. * STUDY DESIGN This is a prospective, longitudinal, within-person (split-mouth) interventional study. Each participant receives both trimming-line designs simultaneously: supragingival on one dental arch and juxtagingival festooned on the contralateral arch. The split-mouth design controls for inter-individual variation in systemic health, oral hygiene, salivary composition, and baseline microbiological status. The aligner material (CA Pro tri-layer, SCHEU-DENTAL GmbH, Iserlohn, Germany; CE-marked thermoplastic) is identical for both designs; only trimming-line geometry differs. * ALLOCATION RULE (NON-RANDOMIZED) Allocation follows a pre-specified hierarchical rule documented prior to enrollment: 1. Primary criterion: The dental arch with the greater magnitude of planned dental intrusion (mm, from the digital treatment setup prior to fabrication) receives the supragingival design, consistent with its clinical indication for vertical force transmission. 2. Tiebreaker (equal or zero planned intrusion on both arches): A pre-specified systematic alternating sequence in enrollment order. The first participant meeting this criterion receives supragingival on the maxillary arch; the second on the mandibular arch; and so on. This ensures approximate 50/50 balance in arch-level distribution and prevents arch-level confounding. Planned intrusion magnitude per arch is recorded as a continuous covariate in all statistical models. \- BASELINE ASSESSMENT (T0) A single pooled baseline specimen is collected by sampling three index sites per arch (six paper points total: three per arch), combined into one Eppendorf tube at T0. This is justified by the fundamental assumption of the split-mouth design: in participants with clinically healthy periodontal status at baseline (inclusion criterion: probing depth 3 mm or less, no BOP at index sites), no clinically meaningful arch-specific microbiological difference is expected prior to any intervention. The subgingival microbiome of periodontally healthy individuals reflects the common salivary reservoir shared by both arches. The T0 specimen characterizes overall baseline pathogen status. The primary between-arm comparison is conducted at T1 and T2 only, where arch-specific specimens are available. \- SPECIMEN TOTALS: 175 PCR tubes (35 evaluable participants x 5 tubes per participant: 1 pooled at T0 + 2 arch-specific at T1 + 2 arch-specific at T2). Total paper points: 525 (35 x 15: 6 at T0 \[3 per arch x 2 arches pooled\] + 3 at T1 per arch x 2 arches + 3 at T2 per arch x 2 arches). The unit of PCR analysis remains one result per arch per time point; pooling of three paper points per arch increases the biological material per tube and improves PCR sensitivity without changing the analytical unit or the power calculation. \- SAMPLE COLLECTION PROCEDURE All collections are performed under standardized fasting conditions (a jeun): participants refrain from eating, drinking (water permitted), brushing, using mouth rinse, and from removing the aligner for at least 8 hours prior to collection (overnight aligner wear). Collections take place in the morning. At each visit, supragingival plaque is gently removed from each index tooth with a sterile curette without entering the sulcus, and the area is isolated with cotton rolls. Three sterile paper points (size #25, as specified by the Sacace PeriodontScreen Real-TM kit) are sequentially inserted into the three index sites per arch, each held in place for 10-20 seconds. The three paper points from the same arch are immediately pooled into a single Eppendorf tube (200 ul sterile PBS), yielding one biological specimen per arch per time point. \- INDEX SITE SELECTION (pre-specified, recorded at T0, unchanged at T1 and T2): Primary rule: The three sites with the greatest probing depth within each arch are selected as index sites, regardless of their anatomical zone (anterior, premolar, or molar). All three sites may be in the same zone if those are the deepest. Tiebreaker - Maxillary arch (when two or more sites share the greatest probing depth and fewer than three unique deepest sites can be distinguished): mesial of tooth 1.6 + mesial of tooth 2.6 + central of tooth 1.1. This selection is consistent with the maxillary sampling sites used by Lombardo et al. (2021), who collected subgingival specimens from the right upper first molar and right upper central incisor. Tiebreaker - Mandibular arch: mesial of tooth 4.6 + mesial of tooth 3.6 + central of tooth 4.1. This rule is defined a priori by the investigator as the anatomical mirror of the maxillary tiebreaker (molar-molar-incisor), applied to the mandibular arch using the same tooth types in the corresponding positions. No published study on subgingival microbiota in orthodontic patients was identified that used an identical mandibular tiebreaker rule; the rule is therefore pre-specified and documented here as an investigator-defined protocol decision. The same three index sites per arch are used at T1 and T2. \- COMPLIANCE MONITORING AND DROPOUT PROTOCOL Aligner wear compliance (target 22 hours/day or more) is assessed by self-report at each study visit. Participants who report compliance below 22 hours/day at two or more consecutive visits are withdrawn from the study and their data are excluded from per-protocol analysis. A single visit with reported non-compliance triggers verbal reinforcement and is recorded as a protocol deviation. All enrolled participants are included in an intent-to-treat description regardless of compliance status. At T0: six paper points (three per arch) are all pooled into one Eppendorf tube (200 ul sterile PBS) - one tube total per participant at baseline. At T1 and T2: three paper points per arch are pooled into separate Eppendorf tubes per arch (supragingival-design arch and juxtagingival-design arch), labeled with participant code, time point, and arch designation - two tubes per participant per time point. Tubes are frozen at -20 degrees C within 2-4 hours at Aorys Clinic, Sibiu. Samples are transported in batches to the Department of Microbiology, UMFST Targu Mures, in insulated containers with frozen gel packs (transport approximately 2-2.5 hours). Storage at -20 degrees C before analysis is consistent with validated protocols for subgingival bacterial DNA preservation. \- LABORATORY ANALYSIS Samples are analyzed at the Department of Microbiology, George Emil Palade University of Medicine, Pharmacy, Science, and Technology of Targu Mures and Mures County Clinical Hospital, using the PeriodontScreen Real-TM kit (Sacace Biotechnologies Srl, Via Scalabrini 44, 22100 Como, Italy; REF T01707-96-S; CE-marked IVD). DNA extraction: Paper points are vortexed in PBS (30 seconds) to elute bacteria. Suspension is centrifuged (12,000 x g, 2-3 minutes), supernatant removed, and bacterial DNA extracted from the pellet using a commercial extraction kit (QIAamp DNA Mini Kit, Qiagen, or equivalent validated kit) per manufacturer instructions. Real-time PCR: The PeriodontScreen Real-TM assay detects and quantifies seven periodontal pathogen species using specific fluorescent probes (FAM channel, 40 cycles; HEX channel internal control). Kit-validated clinical significance thresholds: A. actinomycetemcomitans 10\^4 copies/reaction or more; P. gingivalis, T. forsythia, T. denticola 10\^5 or more; P. endodontalis, F. nucleatum, P. intermedia 10\^6 or more. A result is positive for a given pathogen when FAM Ct is less than 35. A result is considered valid when FAM Ct is less than 35 (regardless of HEX Ct value), or when HEX Ct is less than 35. A result is considered invalid and must be repeated when both FAM Ct and HEX Ct are greater than 35 or absent, indicating possible extraction failure or PCR inhibition. Laboratory personnel receive coded tubes (participant code, arch, time point) without knowledge of which arch corresponds to which trimming-line design. Additionally, the sample collector (a general dentistry specialist and orthodontic resident, separate from the treating clinician) is completely blinded to arch allocation at all time points: the collector performs all subgingival sampling without knowledge of which trimming-line design is assigned to each arch. The treating clinician (PI) retains the allocation key, released only after all PCR analyses are complete. \- STATISTICAL ANALYSIS Confirmatory primary analysis: The confirmatory endpoint is the arch-level pathogen burden score (0-7) at T2 (8 weeks). A single Wilcoxon signed-rank test (paired, two-sided, alpha = 0.05) will compare supragingival versus juxtagingival arch scores at T2. This test was selected because the outcome is a discrete ordinal count (integers 0-7) in a paired split-mouth design, for which non-parametric paired comparison is appropriate and does not require distributional assumptions. A single confirmatory test at T2 is pre-specified to control the type I error rate at alpha = 0.05 without correction for multiplicity. \- Supportive analysis at T1: The Wilcoxon signed-rank test will also be applied at T1 (4 weeks) as a supportive, non-confirmatory analysis to characterize the temporal pattern of any observed effect. Results at T1 will be reported descriptively; no confirmatory inference will be drawn from the T1 test. Adjusted analysis: A Generalized Estimating Equations (GEE) model will be fitted to arch-level pathogen burden scores at T1 and T2, with the following specification: (1) response: arch-level pathogen burden score (0-7); (2) link function: identity; (3) distribution family: Gaussian; (4) working correlation structure: exchangeable, appropriate for two post-baseline time points with equal within-subject correlation assumed across time; (5) covariates: trimming-line design (binary: supragingival vs. juxtagingival), time point (categorical: T1, T2), planned intrusion magnitude per arch (continuous, mm). The design x time point interaction term will be tested to assess whether the effect of trimming-line design differs between T1 and T2; if the interaction term is not statistically significant (p \> 0.10), it will be dropped and a main-effects-only model will be reported. The GEE model accounts for within-subject correlation across time points and adjusts for the pre-specified non-random allocation variable (planned intrusion magnitude). GEE analyses will be performed using the geepack package in R (version 4.0 or later). * Exploratory analysis: McNemar's test comparing arch-level red complex positivity (binary outcome) between the two designs at T1 and T2. Interpreted descriptively only; no confirmatory inference will be drawn. Anticipated power for this exploratory outcome is approximately 22%, due to the low expected prevalence of simultaneous red complex positivity in periodontally healthy participants. This low power is acknowledged and the outcome is registered as exploratory accordingly. * Missing data: Missing outcome data will be handled by complete-case analysis. Participants with evaluable data at both T1 and T2 will constitute the per-protocol population used in the confirmatory analysis. All enrolled participants, including those who withdraw after T0, will be described in the intent-to-treat population; no imputation will be performed for the primary analysis, given the anticipated dropout rate of 10% or less. * Software: Primary confirmatory analysis (Wilcoxon signed-rank test) and descriptive statistics will be performed using GraphPad Prism 8 (version 8.0.2). GEE models will be performed in R (version 4.0 or later). All reported p-values are two-sided. No interim analyses are planned. * SAMPLE SIZE JUSTIFICATION Sample size was determined by Monte Carlo simulation (10,000 iterations) for a paired within-subject (split-mouth) design. Simulation parameters were selected conservatively in the absence of published pilot data specifically on trimming-line design effects on subgingival pathogen burden: (1) minimum clinically meaningful difference between arches: 0.5 species above threshold, representing one additional pathogen species crossing the clinical significance threshold in the supragingival compared to the juxtagingival arch; (2) standard deviation of individual arch-level scores (sigma): 0.88, implying a standard deviation of paired differences (sigma\_D) of 0.964 \[sigma\_D = sigma x sqrt(2 x (1 - rho)) = 0.88 x sqrt(1.2) = 0.964\]; this represents a conservative estimate of between-participant variability in the absence of pilot data; paired differences were simulated as approximately symmetric (Gaussian copula with normal marginals), which is a conservative baseline for Wilcoxon power estimation; (3) intra-subject correlation rho = 0.4, consistent with values reported for split-mouth periodontal microbiological studies; (4) two-sided alpha = 0.05; (5) Wilcoxon signed-rank test. Simulation yielded estimated power = 82.7% (primary) and 82.3% (independent verification), with empirical type I error = 5.0%, at n = 35 evaluable participants. To account for an anticipated dropout rate of approximately 10% (loss to follow-up, non-compliance with 22 h/day aligner wear, protocol deviations), an enrollment target of 39 participants was set. \- LIMITATIONS The PeriodontScreen Real-TM kit was validated for populations with active periodontal disease; its clinical significance thresholds may not be fully calibrated for the healthy periodontal status of enrolled participants. Results will be interpreted accordingly. The non-randomized allocation rule introduces potential for arch-level biomechanical confounding beyond planned intrusion magnitude (e.g., torque corrections, transverse movements), which are not fully captured by the GEE covariate. Planned intrusion magnitude per arch is recorded and adjusted for as a continuous covariate in the GEE model; residual biomechanical confounding from other movement types cannot be excluded. Because the T0 specimen represents a pooled bilateral sample, arch-specific baseline pathogen scores are not available. The primary analysis therefore compares arch-specific scores at T1 and T2 directly, without adjustment for individual arch-level baseline values. This limitation is mitigated by the inclusion criterion of clinically healthy periodontal status at baseline (probing depth 3 mm or less, no bleeding on probing at index sites), which minimizes the probability of pre-existing arch-level microbiological asymmetry prior to intervention.