The reconstruction of aesthetics and masticatory function using implants can be performed predictably in a variety of clinical situations. However, this treatment cannot always be performed in a favorable oral environment^1,2). For example, implantation in cases of seriously advanced bone loss remains a surgical and prosthetic challenge for clinicians. The limited availability of bone tissue hampers or even prevents implant treatment without the previous application of regenerative techniques³⁾.

Numerous advances in surgical techniques have been achieved to address these challenges, and many have improved the quality of implant and prosthetic treatments. Sinus elevation, guided bone regeneration (GBR), onlay bone grafting, distraction osteogenesis, and displacement of the inferior alveolar nerve have been developed and applied frequently when implant placement is not possible. However, these techniques have several limitations, including limited indications and increased operating times, costs, and complication rates^4-7). Moreover, their application may preclude the use of provisional dentures and reduce success rates⁸⁾.

As an alternative to complex surgeries entailing many of these problems, the use of short dental implants should be considered. Along with their simplicity, these implants allow for less expensive and faster treatment with reduced morbidity^9,10). With the development of new implant surfaces and designs, the use of short implants has increased.

Prospective clinical trials and systematic reviews have shown that short implants have high survival rates, with a low incidence of prosthetic and biological complications^11-15). However, implant survival is a minimal criterion because functional status can be achieved under conditions ranging from progressive bone loss to the absence of such loss. Even with severe peri-implant marginal bone loss (MBL), an implant can continue to function without mobility for several years; such bone loss, however, may lead to the development of deep pockets or recession causing implant surface exposure and patient discomfort. Thus, the maintenance of marginal peri-implant bone is a fundamental prerequisite for long-term implant success¹⁶⁾.

Recent clinical trials have shown that the degree of MBL occurring around short implants is similar to that occurring around implants of conventional length^17,18). However, these results must be interpreted with caution, as the same amount of MBL can be more harmful in the former case. Such loss changes the clinical crown-to-implant ratio (CIR), which can increase the tension on bone tissue around the most cervical region of an implant¹⁹⁾. Thus, the examination and prevention of factors contributing to MBL is very important. This study was conducted to assess MBL and related risk factors associated with the use of dental implants with shorter than conventional lengths placed in the posterior regions of the mandible and maxilla.

The Institutional Review Board of Yonsei University approved the retrospective chart review and data collection performed for this study (#3-2022-0041) (approval date: 2022.03.22).

1. Patient selection

This retrospective cohort study was conducted with clinical and radiographic follow-up periods of up to 59 months. Patients were selected for inclusion using the following criteria:

1)receipt of treatment with short implants (7-mm length, 4-mm diameter; AnyOne^®; MegaGen, Daegu, South Korea);

2)performance of all surgical and restorative treatments by periodontist (D.W.L) at the Department of Periodontology, Gangnam Severance Dental Hospital, Seoul, Korea;

3)≥1 year follow-up after functional implant loading and performance of standardized periapical radiography at the time of final restoration delivery (baseline) and 12 months after functional loading;

4)ability to understand and perform oral hygiene maintenance procedures;

5)good general health; and

6)age ≥20 years.

The exclusion criteria were untreated active periodontitis, American Society of Anesthesiologists score ≥3, bruxism/parafunction, heavy smoking (>10 cigarettes per day), poor oral hygiene (modified plaque index >2), use of a removable partial denture or complete denture in the opposing arch, history of the use of a medication associated with bone metabolism (e.g., bisphosphonate), sinus pathology (e.g., osteoma, carcinoma, other cancer), postoperative complication, uncontrolled compromised systemic disease, and unsuitable head position on a radiograph or poor image quality.

2. Surgical procedure

The treatments followed routine surgical and clinical protocols. All surgical procedures were conducted using a two-stage treatment protocol. Implant placement was performed under local anesthesia (2% lidocaine, 1:100,000 epinephrine; Yuhan Corp., Seoul, Korea) with a linear mid-crestal incision and full-thickness flap, as well as a releasing incision when necessary. With a 20:1 contra-angle (975 AE; W&H Dentalwerk Bürmoos, Salzburg, Austria) attached to a surgical unit (W&H ImplantMED SI-95 230 Power Unit; W&H Dentalwerk Bürmoos, Salzburg, Austria), the bone site was drilled at a speed of 900 rpm with the sequence of burs recommended by the manufacturer up to the diameter suitable for the placement of a regular-neck implant. The implant was inserted at a speed of 18 rpm up to the junction of the rough surface and the smooth collar. Then, 4.0 nylon sutures were placed. All patients were prescribed antibiotics and anti-inflammatory agents. The sutures were removed 7∼10 days after surgery. Second operations in the mandible and maxilla were performed 3 and 6 months, respectively, after the initial operations. Prostheses were delivered 3 weeks after the second operations. The patients were recalled every 6 months for plaque control and oral hygiene management.

3. Radiological evaluation

Periapical radiographs (70 kVp, 8 mA, Planmeca Intra, Helsinki, Finland) were taken on image plates (Vistascan Image Plate, Dürr Dental) immediately after functional loading, 6 months after prosthesis delivery, and at follow-up visits with XCP (XCP^®, Dentsply Sirona) using a parallel cone technique, and scanned using an image plate scanner (Vistascan Perio Plus, Dürr Dental) (Fig. 1). For the measurement of soft-tissue thickness, a radiopaque material (Vitapex^®; Neo Dental Chemical Products Co. Ltd, Tokyo, Japan) was placed on top of the tissue. The X-ray tube was positioned at a 90° angle to the long axis of the dental implant to ensure that the implant threads were visible on mesial and distal images. The scanned files were transferred to a computer (Intel i5-7600 3.5GHz, Santa Clara, CA, USA; Windows 10 Enterprise, Redmond, WA, USA) and the measurements of periapical radiographs were made in a dark room using the same monitor (LS24C45KBSS, Samsung, Seoul, Korea). Measurements of MBL and CIR were made using ImageJ software (1.52a; National Institutes of Health, Bethesda, MD, USA). Calibration was performed using the known diameter of a spherical metal bearing (5.5 mm). Changes in the peri-implant marginal bone level were then measured on radiographs taken at baseline and up to 59 months after prosthesis delivery. The margin between the polished and rough surfaces of each implant was defined as the reference point, and the distances between this point and the most apical points of the marginal bone on the mesial and distal surfaces were measured. These values were then averaged. The same examiner (H.S.L) performed all radiographic measure-ments. To test intra-observer variability, MBLs on 30 randomly selected radiographs which were not included in the present study, were measured twice with a 1-week interval. Pearson’s correlation coefficient was calculated to analyze the correlation between the two sets of measurements. Correlation of the two measurements was significant (Pearson’s correlation coefficient=0.99; P<0.01). The intra-observer variability and correlation coefficient were comparable to previous studies²⁰⁾.

Figure 1. Example of periapical radiography; periapical radiographs taken (A) on the day of functional loading and (B) 6, (C) 12, (D) 18, (E) 24, (F) 30, (G) 36, and (H) 42 months after prosthesis delivery.

4. Statistical analysis

Statistical analyses were performed using SPSS software (version 28; IBM Corporation, Armonk, NY, USA). All results are provided as means±standard deviations. After collecting all data, data distribution normality was tested using Kolmogorov-Smirnov test. As the normality of the distribution was rejected, Mann-Whitney test was used for comparisons of MBL according to history of GBR procedure and prosthesis type (single/splinted). Simple linear regression analysis was used for comparisons of MBL according to soft-tissue thickness and CIR. Multiple linear regression analysis was performed to evaluate the effects of various predictors on MBL. The significance level was set to P<0.05.

1. Study population

In total, 105 implants in 78 patients (42 males and 36 females; mean age, 59±23 years) were included in this study. Information on patients’ demographic characteristics and the surgeries and prostheses is provided in Table 1 and 2. The mean follow-up period was 22.3±9.7 months (range, 12∼59 months; Table 3).

Table 1 . Variables investigated at the implant and patient levels

Implant		Patient
Site		Sex
Maxillary posterior	50	Male	42
Mandibular posterior	55	Female	36
History of GBR procedure		DM
Yes	65	Yes	12
No	40	No	66
Prosthesis type		Cardiovascular disease
Single	40	Yes	33
Splinted	65	No	45
		Smoking
		Yes	15
		No	63

GBR: guided bone regeneration, DM: diabetes mellitus.

Table 2 . Distribution of the installed implants (n=105)

Jaw	Placed site														Total
	7	6	5	4	3	2	1	1	2	3	4	5	6	7
Maxilla	2	9	9	4								5	14	7	50
Mandible	7	13	6	1							4	5	12	7	55

Table 3 . General characteristics of dental implants (n=105)

Variables	Mean±SD	Min-Max
Follow-up period (month)	22.3±9.7	12∼59
Mean MBL (mm)	0.08±0.13	0.0∼0.48

SD: standard deviation, MBL: marginal bone loss.

2. Marginal bone loss and survival rate

The cumulative mean amount of peri-implant bone loss during follow-up was 0.08±0.13 mm (range, 0.0∼0.48 mm; Table 3). The cumulative implant survival rate was 100%.

3. Risk factors

No significant increase in the risk of MBL according to history of GBR procedure or prosthesis type was detected (Table 4, Fig. 2). Simple linear regression analysis revealed no significant increase in the risk of MBL according to the CIR. The soft-tissue thickness was an MBL risk factor (Table 5, Fig. 3). The multiple linear regression analysis yielded no significant result for any potential MBL risk factor (Table 5).

Table 4 . Amount of marginal bone loss according to history of GBR procedure and prosthesis type (n=105)

Variables	Category	Subjects	MBL (mm)	P-value
History of GBR procedure	Yes	65 (61.9)	0.08±0.13	0.586
	No	40 (38.1)	0.08±0.12
Prosthesis type	Single	40 (38.1)	0.09±0.14	0.775
	Splinted	65 (61.9)	0.07±0.12

Data are presented as n (%) or mean±SD.

GBR: guided bone regeneration, MBL: marginal bone loss.

Table 5 . Estimated risk of marginal bone loss according to various factors

	Univariable model		Multivariable model
	Beta (SE)	P-value	Beta (SE)	P-value
Prosthesis type	−0.016 (0.025)	0.524	−0.019 (0.025)	0.458
Soft tissue thickness	−0.028 (0.014)	0.048*	−0.023 (0.016)	0.164
History of GBR procedure	0.006 (0.025)	0.812	−0.004 (0.026)	0.879
CIR	−0.061 (0.034)	0.079	−0.033 (0.040)	0.409

*P<0.05.

GBR: guided bone regeneration, CIR: crown-to-implant ratio.

Figure 2. Box plot of marginal bone loss according to guided bone regeneration use and prosthesis type. The lower and upper box limits represent the 25th and 75th percentiles, respectively.

Figure 3. Scatter plot of marginal bone loss in relation to soft-tissue thickness. The regression line is given as dotted line.

In this retrospective cohort of 105 short implants, most peri-implant regions showed only slight (∼0 mm) bone loss, and the incidence of pathological MBL was very low. Derks and colleagues²¹⁾ analyzed national social insurance agency data on bone loss in randomly selected patients who underwent prosthetic treatment and reported a mean value of 0.72±1.15 mm at about 9 years, much greater than the value obtained in this study. In contrast to the present study, various implant systems were included in that study, and general practitioners treated 74% of the included patients. In addition, our patients received regular supportive periodontal treatment performed by a periodontal specialist at each visit for the maintenance of periodontal health, which may have contributed to the minimal changes in the marginal bone level.

Our multiple linear regression analysis revealed no significant MBL risk factor. On the other hand, simple linear regression analysis suggested that the soft-tissue thickness was the only factor affecting MBL around the short implant system examined. Thus, although the use of this system can yield highly predictable long-term results, the soft-tissue thickness needs to be adequate to ensure the most favorable outcomes. Berglundh and Lindhe²²⁾ demonstrated in an animal study that a certain amount of mucosal thickness was needed to establish biological width around a dental implant. In cases of insufficient thickness, crestal bone resorption occurs until sufficient space for the connective tissue and junctional epithelium is created. Although the compositions and structures of natural teeth and implants are similar, the implant attachment apparatus is longer than a natural tooth, meaning that greater soft-tissue thickness is required around implant fixtures^23-25). Linkevicius et al.²⁶⁾ and Puisys and Linkevicius²⁷⁾ demonstrated that significantly less MBL occurred in the presence of thick tissue or augmented thin tissue than in the presence of non-augmented thin tissue. Our finding indicates that the soft-tissue thickness is equally important for short implants.

A limitation of this study is that other various factors can lead to marginal bone loss. Although some factors, such as implant design, loading protocol were controlled, it was not possible to control for some other factors, such as any medical condition, smoking habit, or individual susceptibility to periodontitis and peri-implantitis, which might influence marginal bone loss. Moreover, this study was based on the retrospective analysis of a large sample of data from patients whose follow-up visits were not precisely controlled, which may pose a risk of missing data due to gaps in information or incomplete records. To confirm the prevalence of MBL and to identify risk factors for such loss, a prospective study conducted with a large sample and long-term observational period is needed.

Within the limitations of this study, the findings suggest that insufficient soft-tissue thickness is a risk factor for MBL in patients with short (7-mm-long, 4-mm-diameter) implants. The other factors examined showed no significant relationship to MBL.