Angled Screw Channel‐Retained vs. Cement‐Retained Implant Crowns in Nonmolar Sites: A Systematic Review and Meta‐Analysis

ABSTRACT

Objectives

To evaluate the clinical, radiographic, and aesthetic outcomes of angled screw channel (ASC) retained implant crowns to cement‐retained implant crowns in nonmolar sites.

Methods

Randomized and nonrandomized trials comparing ASC‐retained and cement‐retained implant crowns in single nonmolar locations were found by searching electronic databases (COCHRANE, EMBASE, and MEDLINE) up to January 2025. Changes in marginal bone level were the primary outcomes, whereas periodontal parameters, aesthetic outcomes, and technical complications were the secondary outcomes. Random‐effects meta‐analysis was used to calculate pooled effect sizes.

Results

A total of 973 studies were identified, of which four studies with 167 single nonmolar implant crowns were included. Overall meta‐analysis showed that the difference in marginal bone level was in favor of the ASC group, but the difference was not statistically significant (MD −0.03; 95% CI −0.12 to 0.06; p = 0.57). The ASC group had more positive changes in pink aesthetic score than the cemented group; however, the difference between the two groups was not statistically significant (MD −0.18; 95% CI −0.88 to 0.51; p = 0.61).

Conclusions

ASC‐retained implant crowns in nonmolar sites have comparable short‐term clinical, radiographic, and aesthetic outcomes to cement‐retained implant crowns, with less bleeding on probing expected with ASC‐retained crowns.

Clinical Significance

The superiority of ASC‐retained implant crowns over cement‐retained implant crowns in nonmolars was not proven. There were no significant differences between ASC‐retained and cement‐retained implant crowns in terms of mean changes in marginal bone levels, probing pocket depths, aesthetic outcomes, technical complications, and implant failure.

1. Introduction

The success of implant therapy is not only defined by the lack of implant mobility and marginal bone loss but also by achieving long‐term optimal aesthetic and functional outcomes [1, 2]. Peri‐implant diseases, a bacterial‐induced inflammation of the peri‐implant tissues, remain a common occurrence that can negatively impact the long‐term survival of dental implants [1]. A mean incidence rate of 43% and 22% for peri‐implant mucositis and peri‐implantitis, respectively, was reported [3] with several patient and implant‐related risk factors identified [4, 5, 6, 7]. The influence of prosthetic variables as potential risk factors for peri‐implant diseases has also been recently reported. For example, prosthesis design features such as emergence profile and emergence angle, the retention mechanism, implant‐abutment connection, and restorative material, crown‐to‐implant ratio, and the splinting of adjacent implants have shown variable levels of association with peri‐implant diseases [8, 9, 10, 11, 12].

The selection of the appropriate retention mechanism for an implant‐supported prosthesis is an essential step in the implant treatment planning. Screw and cement retention are the two main retention mechanisms commonly employed [13, 14, 15, 16]. Screw retention offers several advantages, such as ease of prosthesis retrieval and better access to oral hygiene, but requires a prosthodontically driven implant placement for optimal aesthetics [13]. In the anterior region of the maxilla, where the buccal bone is either thin or absent [17, 18], this could be very challenging. Attempting to utilize a screw‐retained prosthesis in this situation would result in the screw‐access hole being placed labially, thereby compromising the aesthetics or the integrity of the prosthesis [19, 20, 21]. On the other hand, the use of cement‐retained prostheses, while aesthetically superior and more forgiving when implant positioning is sub‐optimal, has been associated with higher risks of marginal bone loss and peri‐implant diseases [16, 22]. The difficulty in the complete removal of residual cement and the inaccessibility of the cement‐retained prostheses to oral hygiene measures, with potentials for inflammatory responses, have been well described [22, 23, 24, 25].

One method allowing clinicians to utilize screw‐retained prostheses in anatomically challenging sites is the use of implant abutments with angled screw channels (ASCs) [26]. One early report on ASC described the use of a dynamic abutment system consisting of a metal base with a castable plastic cylinder that can be angled up to 28° relative to the implant's long axis [27]. A special screwdriver with a hexagonal sphere head is then used to tighten the abutment‐crown assembly onto the implant. A recent system of a titanium‐based abutment with ASC that allows the placement of the screw access hole anywhere between 0° and 25° off the implant's original axis is now commonly utilized [28, 29]. The retentive screw of this ASC abutment can be torqued onto the implant using a unique omnigrip screwdriver. ASC abutments allow flexibility in implant placement and offer significant aesthetic advantages as they help in avoiding buccal screw access by positioning the screw access hole in a less visible area. This is particularly beneficial in the anterior maxilla, where maintaining the natural appearance of the restoration is paramount [30]. Moreover, screw‐retained restorations with ASC abutments are easy to retrieve for maintenance or repair [31]. Positive outcomes of implant prostheses with ASC, in terms of function and aesthetics, have been reported in several in vitro and short‐term clinical studies [32, 33, 34, 35, 36, 37, 38].

Systematic review on ASC implant prostheses has also presented acceptable technical behavior and clinical outcomes [26, 39]. In the majority of these reports, however, comparative control groups or quantitative analyses of treatment outcomes to prove clinical efficacy were lacking [26, 33, 36, 37, 38, 39]. Therefore, this systematic review and meta‐analysis aimed to evaluate clinical, radiographic, and aesthetic outcomes of ASC‐retained implant crowns in comparison to cement‐retained implant crowns in nonmolar sites.

2. Materials and Methods

The current systematic review was prepared in accordance with the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta‐analyses [40] and the Cochrane Collaboration [41]. The PICOS (participant, intervention, comparison, outcome and study design) framework [41, 42] was used to determine the eligibility requirements:

Participant: Human adults over the age of 18 who needed dental implant therapy after having one nonmolar tooth extracted.

Intervention: ASC‐retained implant crowns.

Comparison: cement‐retained implant crowns.

Outcomes: changes in marginal bone level, periodontal parameters, aesthetic outcomes, the necessity for augmentation during implant insertion, technical complications, and implant failure rate.

Study design: randomized and nonrandomized studies with a minimum of 12‐month follow‐up period.

Research question: In human adults over the age of 18 who have had one nonmolar tooth extracted and require dental implant therapy, how do ASC‐retained implant crowns compare to cement‐retained implant crowns in terms of changes in marginal bone level, periodontal parameters, aesthetic outcomes, the necessity for augmentation during implant insertion, technical complications, and implant failure rates over a minimum follow‐up period of 12 months in both randomized and nonrandomized studies?

The study was registered with the National Institute for Health Research (NHR) under the PROSPERO ID CRD42024537973. Ethical approval was not required for this systematic review. This review included randomized and nonrandomized studies comparing screw‐retained implant crowns with ASC to cement‐retained implant crowns in nonmolar sites. The included studies must report on changes in radiographic bone dimensions, periodontal parameters, aesthetic outcomes, the need for augmentation, technical complications (for example: porcelain fracture, loss of retention, screw loosening, and loss of access sealing material), or implant failure (defined as implant loss) rate. Neither publishing status nor language limitations were used. Case reports, case series, in vitro research, and studies that did not provide sufficient data were excluded. Participants were 18 years of age or older and required the replacement of a single missing nonmolar tooth with either an ASC‐retained or a cement‐retained implant crown. The intervention group involved the fabrication of a screw‐retained implant crown with ASC that permitted a screw‐access angle of up to 25°, whereas in the control group, a cement‐retained implant crown was fabricated. Primary outcomes measured included changes in marginal bone level, while secondary outcomes assessed changes in bleeding on probing, changes in probing pocket depth, aesthetic outcomes (pink aesthetic score (PES) and white aesthetic score (WES)), the necessity for augmentation during implant insertion, technical complications, and the implant failure rate.

2.1. Search Strategy and Selection of Studies

Standard search procedures were followed [41, 43]. The Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE, MEDLINE, and ClinicalTrials.gov were searched independently and twice by the two review authors (M.A.A. and N.H.M.A.) for published, continuing, and unpublished studies until January 12, 2025. The search keywords included a combination of MeSH and non‐MeSH terms that were relevant to the research question and the scope of the present systematic review (Table 1). Furthermore, bibliographies of each chosen paper were completed, along with a manual search of additional research published in relevant dental journals over the previous 5 years (Journal of Aesthetic and Restorative Dentistry, Clinical Implant Dentistry and Related Research, Clinical Oral Implants Research, International Journal of Oral and Maxillofacial Implants, International Journal of Periodontics and Restorative Dentistry, International Journal of Prosthodontics, Journal of Clinical Periodontology, Journal of Periodontology, and Journal of Prosthodontics).

TABLE 1.

Databases and search terms.

Databases	Keywords
Published studies
PubMed (1965—January 12, 2025) (n = 1009)	(Angled screw channel OR cement retained) AND (dental implant OR oral implant) AND (screw retained OR cement retained)
EMBASE via Ovid (1947—January 12, 2025) (n = 553)	(Angled adj screw adj channel).mp OR (cement adj retained).mp. AND (dental adj implant).mp. OR (oral adj implant).mp. AND (screw adj retained).mp. OR (cement adj retained).mp.
Cochrane Central Register of Controlled Trials (CENTRAL) via Ovid (January 12, 2025) (n = 58)	(Angled adj screw adj channel).mp OR (cement adj retained).mp. AND (dental adj implant).mp. OR (oral adj implant).mp. AND (screw adj retained).mp. OR (cement adj retained).mp.
Unpublished studies
ClinicalTrials.gov (January 12, 2025) (n = 0)	(Angled screw channel OR cement retained) AND (dental implant OR oral implant) AND (screw retained OR cement retained)

Two review authors (M.A.A. and N.H.M.A.) checked the retrieved citations separately and in duplicate. This initial screening was based on the title, abstract, and keywords of the citations to identify potentially relevant studies. Any citations deemed irrelevant were excluded at this stage. For the remaining citations, the full texts of the papers were obtained and further scrutinized using a predefined eligibility form to ensure they met the inclusion criteria. Throughout this process, any disagreements between the two review authors were resolved through discussion and consensus. If consensus could not be reached, a third review author (M.S.) was consulted to make the final decision. In cases where multiple publications of the same study were identified, the publication containing the most comprehensive and relevant information was selected for inclusion. Additionally, the reasons for excluding studies at each stage of the selection process were meticulously documented and reported to maintain transparency and reproducibility of the review.

2.2. Data Collection and Quality Assessment of Included Studies

Using a data extraction form, the two review writers (M.A.A. and N.H.M.A.) independently gathered the following data from the included studies: (1) Features of the study: title, authors, location, language, year, published or unpublished data, funding source, study design (split mouth or parallel group), randomization technique, allocation concealment, and blinding (participants, investigators, outcome examiners). (2) Participants: attrition rate, reasons for dropouts, number of participants in test and control groups, demographics, and inclusion/exclusion criteria. (3) Interventions: the number of patients or implants that received a single ASC‐containing implant crown. (4) Comparison: the quantity of subjects or implants that employed a single cement‐retained implant crown. (5) Results: variations in the depth of the probing pocket, variations in bleeding during probing, variations in the marginal bone level, aesthetic results, the requirement for augmentation after implant placement, technical complications, and implant failure rate. (6) The observation period's duration. A third review author (M.S.) was consulted or a consensus was reached to settle any disputes amongst the reviewers. The respective authors of the included studies were contacted to request missing data or obtain more information if needed. Two reviewers (M.A.A. and N.H.M.A.) independently and in duplicate evaluated the risk of bias for each of the included studies. The randomized and nonrandomized trials were evaluated using the Cochrane Collaboration's Risk of Bias methods, namely RoB2 and ROBINS‐I [41].

2.3. Data Synthesis

Meta‐analyses for studies of similar comparisons reporting the same outcome measures were conducted using a statistical software program (Review Manager (RevMan) software, version 5.3, The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark). For instance, 95% CIs and mean difference (MD) or standardized mean difference (SMD) were employed to characterize continuous data, including variations in marginal bone level. A random‐effects model was used to aggregate the findings from several research studies because heterogeneity between them was expected.

Since the ability to identify publication bias was limited when fewer than 10 papers were included, publication bias was not evaluated [41]. The statistical heterogeneity between several studies was assessed using the I ² statistic and Cochran's test for heterogeneity [41]. When the I ² value exceeded 60, there was significant heterogeneity. The statistical unit of analysis was regarded as the participant. Studies and participants with complete data were included, and missing data were assumed to be entirely missing at random. To examine the origins of heterogeneity, the stability of the findings, and the impact of the studies, a leave‐one‐study‐out sensitivity analysis was conducted. Sensitivity analysis was performed to ensure the robustness of our findings and determine whether estimated effects varied when studies with a high probability of bias were excluded from the study. The degree of evidence certainty was assessed using the five GRADE criteria (risk of bias, inconsistency, imprecision, indirectness, and publication bias) [41]. The summary of findings table was created using the GRADEpro Guideline Development Tool software, which is accessible via gradepro.com and was developed by McMaster University and Evidence Prime in 2021.

3. Results

3.1. Characteristics of the Study Settings

The databases yielded a total of 1581 studies (Figure 1). Eleven studies qualified for full‐text evaluation after titles and abstracts were reviewed independently and twice by two review writers (M.A.A. and N.H.M.A.) [28, 29, 31, 33, 36, 38, 44, 45, 46, 47, 48]. Five studies [31, 33, 36, 38, 47] were excluded because they did not include a control group, one was a duplicate [44] and another was an in vitro investigation [46]. As a result, four studies [28, 29, 45, 48] were included in the present review (Table 2). No additional studies were found through the manual search. Of the four included studies, three were conducted in China [28, 29, 48], and one in Italy [45]. All the included studies were conducted in a university setting, and all were of a parallel‐group design. One study [48] was funded by a university research grant, another [45] was self‐funded, while the remaining two [28, 29] did not disclose their source of funding.

FIGURE 1.

Flowchart of the search process.

TABLE 2.

Characteristics of the included studies.

	Lv et al. (2021)	Nastri et al. (2021)	Shi et al. (2020)	Yang et al. (2023)
Study design	RCT (parallel group)	NRSI	NRSI	NRSI
Location	Shanghai Jiaotong University, Shanghai, China	University of Campania Luigi Vanvitelli, Naples, Italy	Shanghai Jiaotong University, Shanghai, China	Zhejiang University, Zhejiang Province, China
Number evaluated (participants/implants)	56/56	20/20	43/43	48/48
ASC‐retained	29/29	10/10	23/23	28/28
Cement‐retained	27/27	10/10	20/20	20/20
Age (years)	31.0 ± 9.8	47.5 ± 10.5	33.9 ± 12.3	38.7 ± 13.0
Smoking habits	NR	None	NR	NR
Implant system	a	a	a	a , b
Implant diameter (mm)	3.5	NR	3.5	3.0–4.3
Timing of implant placement [49]
ASC‐retained	Type I: 8	Type IV: 10	Type I: 23	Type IV: 28
	Type IV: 22
Cement‐retained	Type I: 7	Type IV: 10	Type I: 20	Type IV: 20
	Type IV: 23
Types of crown
ASC‐retained	One piece zirconia coping veneered with ceramic c	NR	One piece zirconia coping veneered with ceramic c	NR
Cement‐retained	Two‐piece zirconia‐based restoration d veneered with ceramic c and cemented e onto prefabricated titanium abutment	NR	Two‐piece zirconia‐based restoration d veneered with ceramic c and cemented f onto customized zirconia abutment	NR
Mean angulation of the screw channel	NR	NR	13.7°	NR
Implant location	Maxillary nonmolar teeth	Maxillary and mandibular nonmolar teeth	Maxillary nonmolar teeth	Maxillary nonmolar teeth
Methods of assessment	Periapical radiographs with paralleling technique g , periodontal probe h , intraoral photos	Periapical radiographs, periodontal probe, intraoral photos	Digital periapical radiographs with paralleling technique g , periodontal probe h	Periodontal probe i ,CBCT
Changes in MBL (mm)
ASC‐retained	−0.17 ± 0.39	−0.25 ± 0.19	−0.31 ± 0.30	−0.28 ± 0.56
Cement‐retained	−0.19 ± 0.32	−0.29 ± 0.11	−0.41 ± 0.38	−0.15 ± 0.36
BoP (%)
ASC‐retained	21.84 ± 19.97	NR	11.60 ± 19.10	13.18 ± 20.00
Cement‐retained	37.04 ± 26.28	NR	33.30 ± 33.80	28.35 ± 22.92
PPD (mm)
ASC‐retained	2.30 ± 0.68	4.00 ± 0.66	2.31 ± 0.60	1.61 ± 0.20
Cement‐retained	2.50 ± 0.60	3.70 ± 0.82	2.48 ± 0.50	1.70 ± 0.50
PES
ASC‐retained	12.42 ± 1.28	7.7 ± 0.82	8.96 ± 0.88	10.63 ± 3.49
Cement‐retained	12.24 ± 1.69	8.8 ± 0.63	8.98 ± 0.62	9.88 ± 2.11
WES
ASC‐retained	9.29 ± 0.75	8.5 ± 1.35	NR	NR
Cement‐retained	9.40 ± 0.65	8.3 ± 0.94	NR	NR
Need for augmentation N (%)
ASC‐retained	NR	6 (60.0)	23 (100.0) j	20 (71.4)
Cement‐retained	NR	3 (30.0)	20 (100.0) j	16 (80.0)
Technical complications N (%)
ASC‐retained	1 (3.5)	0 (0.0)	2 (8.7)	3 (10.7)
Cement‐retained	0 (0.0)	0 (0.0)	1 (5.0)	2 (10.0)
Implant failure rate N (%)
ASC‐retained	0 (0)	0 (0)	0 (0)	0 (0)
Cement‐retained	0 (0)	0 (0)	0 (0)	0 (0)
Follow‐up period (months)	12	44.3 ± 11.5	12	32

Abbreviations: ASC: angled screw channel; BoP: bleeding on probing; CBCT: cone beam computed tomography; MBL: marginal bone level; NR: not reported; NRSI: nonrandomized study of intervention; PES: pink aesthetic score; PPD: probing pocket depth; RCT: randomized controlled trial; WES: white aesthetic score.

^a Nobel Biocare AB, Gothenburg, Sweden.

^b Straumann AG, Basel, Switzerland.

^c VM9, VITA Bad Säckingen, Germany.

^d Lava Zirconia, 3M ESPE, Seefeld, Germany.

^e RelyXTM U200, #M ESPE, SeeFeld, Germany.

^f Glass Ionomer Cement CX‐plus, Shofu, Tokyo, Japan.

^g XCP instruments, Rinn Corporation, Elgin, IL, USA.

^h 15 UNC/CP‐11.5B screening color‐coded probe, Hu‐Friedy, Chicago, IL, USA.

ⁱ Williams‐Sulcus color‐coded probe, Hu‐Friedy, Chicago, IL, USA.

^j Bio‐Oss, Geistlich Pharma AB, Wolhusen, Switzerland.

The baseline characteristics of participants in the selected studies included adults aged ≥ 18 years [28, 29, 48], patients requiring replacement of a single missing nonmolar tooth with either an ASC‐retained or cement‐retained implant crown [28, 29, 45, 48], buccal bone dehiscence of less than 2 mm [29], buccal bone thickness of more than 2 mm [48], and implants that have been in function for at least a year [28, 29, 48] or two [45]. Exclusion criteria included systemic conditions that may affect implant treatment [28, 29, 48], uncontrolled diabetes mellitus (fasting blood glucose > 7.2 mmol/L, glycosylated hemoglobin > 7%) [28, 29, 48], current use of bisphosphonates to treat malignancy [28, 29, 48], history of radiation therapy in the head and neck region [28, 48], untreated periodontitis (defined as full‐mouth plaque score > 20%, full‐mouth bleeding score > 25%, with probing pocket depth of > 5 mm [28, 48], or full‐mouth bleeding score > 20%, with probing pocket depth of > 5 mm [29]), severe periapical infection [29], smoking > 10 cigarettes per day [28, 29, 48], current pregnancy or a near‐future pregnancy intention [28, 48], and an angle greater than 25° between the implant and restoration axes [28].

The characteristics of the interventions included periodontal treatment preceded by any implant planning, and only participants with a bleeding score of < 10% and probing pocket depths of ≤ 4 mm were included [28]. After preoperative clinical and radiographic assessment, implants with a diameter of 3.3–4.1 mm were placed following delayed [28, 45, 48] or immediate implant placement protocols [28, 29]. For immediately placed implants, teeth were extracted as minimally traumatic as possible, and the integrity of the buccal bone plate was assessed by a periodontal probe before implant placement [28, 29]. Simultaneous guided bone regeneration, using deproteinized bovine bone mineral (Bio‐Oss Spongiosa 0.25–1 mm size, Geistlich Pharma, Wolhusen, Switzerland), was performed to fill the gap between the buccal bone and the implant [28, 29]. A healing cap (5 mm) was placed, and the wound was sealed with a collagen sponge and sutured with interrupted monofilament 5‐0 nylon sutures [28, 29]. In one study [29], participants received pre‐ and postoperative antimicrobial prophylaxis.

Prosthetic procedures commenced three to 6 months following surgery [28, 29, 48]. After implant‐level impressions, an implant crown with an ASC abutment (NobelProcera ASC, Nobel Biocare AB) or a cemented implant crown was fabricated based on participants' requirements [29, 48] or random allocation [28]. Prefabricated abutments were used based on the location of the soft tissue margin [28]. In the ASC group, a crown made up of a zirconia coping veneered with a ceramic layer (VM9, VITA, Bad Sackingen, Germany) on a titanium base abutment with a 25° ASC was connected to the implant using a special screw [28, 29]. On the other hand, in the cemented crowns group, a prefabricated titanium [28] or customized zirconia abutments [29, 45] were used. A zirconia‐based crown (Lava Zirconia, 3M ESPE, Seefeld, Germany) with a ceramic veneer (VM9, VITA, Bad Sackingen, Germany) was then cemented over the abutment [28, 29]. Occlusal adjustment was performed to implement an implant‐protected occlusion [28]. A periapical radiograph was then taken using a paralleling technique and a Rinn film holder (XCP instruments, Rinn Corporation, Elgin, IL) [28].

The primary outcome measure was the changes in marginal bone level as measured by cone beam computed tomography (CBCT), standardized periapical radiographs, and image analysis software [28, 29, 45, 48]. Both the day of the final restoration delivery and a year later, CBCT [48] or periapical radiographs using the paralleling approach were taken [28, 29, 45]. The distance between the restoration margin and the coronal level of the implant‐bone contact was recorded using image analysis software, and the mean value of the mesial and distal sites was used. The calibration reference was the implant length. The secondary outcome measures were changes in bleeding on probing and probing pocket depth as measured by periodontal probe [28, 29, 45, 48], aesthetic outcomes as assessed by digital photographs [28, 29, 45, 48], the need for augmentation as assessed at the time of implant placement [29, 45, 48], technical complications [28, 29, 45, 48], and the implant failure rate [28, 29, 45, 48].

3.2. Risk of Bias

A randomized controlled trial [28] was one of the included studies. The study had a rating of “some risk of bias concerns” overall. Although the data assessors' allocation concealment and blinding techniques were not disclosed, it is likely that the evaluation was unaffected by the intervention's knowledge. All outcomes seem to be detected with a low attrition rate, and there were no variations from the planned interventions. Thus, in the other categories, the study was deemed to have a low risk of bias (Table 3). The three nonrandomized studies [29, 45, 48] had varying risks of bias: two [29, 45] were at moderate risk due to confounding factors, while one [48] was at low due to effective covariate analysis. Participant selection and intervention classification biases were minimal across all studies. Outcome measurement bias was low in Yang et al. [48] due to blinded assessors, but high in the other two studies [29, 45]. None of the studies showed high risk of bias from incomplete data or selective reporting (Table 3). Additionally, two studies [28, 48] discussed sample size calculation and were registered before the start of the study.

TABLE 3.

Assessment of risk of bias of the included studies presented with low (green), moderate (yellow) and high (red) risk of bias.

Risk of bias in randomized trials (RoB 2) tool
Domain	Lv et al. (2021)
Bias arising from randomization process
Bias due to deviations from intended interventions
Bias due to missing outcome data
Bias in measurement of the outcome
Bias in selection of the reported result

Risk of bias in nonrandomized studies of interventions (ROBINS‐I) tool
Domain	Nastri et al. (2021)	Shi et al. (2020)	Yang et al. (2023)
Bias arising from confounding
Bias in selection of participants
Bias in classification of intervention
Bias due to deviation from intended interventions
Bias due to missing outcome data
Bias in measurement of outcomes
Bias in selection of reported results

3.3. Effects of Interventions

The current review included 167 patients with 167 implant crowns in nonmolar locations. Ninety of these were ASC‐retained, and the rest were implant crowns that were cement‐retained (Table 4). Data were reported at the participant/implant level in every study. A statistical software (Review Manager (RevMan) software, version 5.3, The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark) was used to perform meta‐analyses for studies with comparable comparisons reporting the same outcome measures.

TABLE 4.

Summary of findings.

Outcome	Number of studies	Relative effect (95% CI)	Anticipated absolute effects a (95% CI)		Certainty of the evidence (GRADE) b
			Cement‐retained	ASC‐retained
Changes in marginal bone level (mm)	4 studies	Not estimable	The mean ranged across control groups from 0.17 to 0.31	MD 0.03 lower (0.12 lower to 0.06 higher)	⊕ ⊕ ⊕⊝ MODERATE c
Changes in bleeding on probing (%)	3 studies	Not estimable	The mean ranged across control groups from 11.60 to 21.84	MD 16.58 lower (24.35 lower to 8.82 lower)	⊕ ⊕ ⊕⊝ MODERATE c
Changes in probing pocket depth (mm)	4 studies	Not estimable	The mean ranged across control groups from 1.61 to 4.00	MD 0.11 lower (0.27 lower to 0.05 higher)	⊕ ⊕ ⊕⊝ MODERATE c
Pink aesthetic score	4 studies	Not estimable	The mean ranged across control groups from 7.70 to 12.42	MD 0.18 lower (0.88 lower to 0.51 higher)	⊕ ⊕ ⊝⊝ LOW c , d
White aesthetic score	2 studies	Not estimable	The mean ranged across control groups from 8.50 to 9.29	MD 0.07 lower (0.42 lower to 0.27 higher)	⊕⊝⊝⊝ VERY LOW c , d , e
Need for additional augmentation at the time of implant placement	3 studies	RR 1.00 (0.90–1.10)	1000 per 1000	1000 per 1000 (900 greater to 1100 greater)	⊕ ⊕ ⊕⊝ MODERATE c
Technical complications	4 studies	RR 1.44 (0.41–5.05)	1000 per 1000	1440 greater per 1000 (410 greater to 5050 greater)	⊕ ⊕ ⊕⊝ MODERATE c

Note: GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Abbreviations: ASC: angled screw channel; CI: confidence interval; MD: mean difference; RR: risk ratio.

^a The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

^b None of the studies suffered from indirectness or detected publication bias.

^c Downgraded one level due to risk of bias: at least one study had no blinding.

^d Downgraded one level due to inconsistency: substantial heterogeneity was detected.

^e Downgraded one level due to imprecision: the effect estimate is based on two studies.

3.3.1. Changes in Marginal Bone Level

Changes in marginal bone level were documented in four studies [28, 29, 45, 48]. The difference in marginal bone level was in favor of the ASC group, but the difference was not statistically significant (MD −0.03; 95% CI −0.12 to 0.06; p = 0.57; Figure 2a). There was no evidence of heterogeneity (I ² = 0%, Chi² = 1.91, df = 3 (p = 0.59)).

FIGURE 2.

Comparison: angled screw‐retained vs. cement‐retained implant crowns. Primary outcome: (a) changes in marginal bone level. Secondary outcomes: (b) changes in bleeding on probing; (c) changes in probing pocket depth; (d) pink aesthetic score; (e) white aesthetic score; (f) need for augmentation at the time of implant placement; (g) technical complications.

3.3.2. Changes in Bleeding on Probing

Changes in bleeding on probing were described in three studies [28, 29, 48]. The meta‐analysis showed a statistically significant difference between the two groups, with more favorable changes in the ASC group compared to the cemented group (MD −16.58; 95% CI −24.35 to −8.82; p < 0.0001; Figure 2b). There was no evidence of heterogeneity (I ² = 0%, Chi² = 0.46, df = 2 (p = 0.80)).

3.3.3. Changes in Probing Pocket Depth

Changes in probing pocket depth were described in four studies [28, 29, 45, 48]. In contrast to the cemented group, the ASC group showed fewer changes in probing pocket depth. However, the difference was not statistically significant (Figure 2c; MD −0.11; 95% CI −0.27 to 0.05; p = 0.18). There was no evidence of heterogeneity (I ² = 0%, Chi² = 1.95, df = 3 (p = 0.58)).

3.3.4. Pink and White Aesthetic Scores

Pink and white aesthetic scores were described in four studies [28, 29, 45, 48]. The meta‐analysis on the pink aesthetic score showed greater improvements in the ASC group compared to the cemented group, but the difference between the two groups was not statistically significant (MD −0.18; 95% CI −0.88 to 0.51; p = 0.61; Figure 2d). Significant heterogeneity was found (I ² = 71%, Chi² = 10.41, df = 3 (p = 0.02)). The white aesthetic score was described in two studies [28, 45]. The meta‐analysis revealed that the difference between the two groups was negligible (MD −0.07; 95% CI −0.42 to 0.27; p = 0.67; Figure 2e). There was no evidence of heterogeneity (I ² = 0%, Chi² = 0.31, df = 1 (p = 0.57)).

3.3.5. Necessity for Augmentation During the Implant Insertion

The necessity for augmentation during the implant insertion was described in three studies [29, 45, 48]. According to the meta‐analysis, there was no statistically significant difference between the two groups (RR 1.00; 95% CI 0.90 to 1.10; p = 0.93; Figure 2f). There was no evidence of heterogeneity (I ² = 4%, Chi² = 2.07, df = 2 (p = 0.35)).

3.3.6. Technical Complications

Technical complications were described in four studies [28, 29, 45, 48]. The technical problems that were found comprised the following: porcelain fracture [48], loss of retention [29], screw loosening [29], and loss of access sealing material [28]. Although the meta‐analysis revealed that the cemented group experienced fewer incidents than the ASC group, there was no statistically significant difference between the two groups (RR 1.44; 95% CI 0.41 to 5.05; p = 0.57; Figure 2g). There was no evidence of heterogeneity (I ² = 0%, Chi² = 0.31, df = 2 (p = 0.86)).

3.3.7. Implant Failure Rate

The implant failure rate was recorded in all of the trials [28, 29, 45, 48]. Since there were no implant failures, a meta‐analysis was not feasible.

3.4. Sensitivity Analyses

According to the leave‐one study‐out sensitivity analysis, the exclusion of one of the included trials did not significantly alter the overall effect‐size estimate for the changes in marginal bone level (pooled MD range, −0.05 to −0.01) or heterogeneity (I ² 0%) Additionally, there was no significant difference between the two groups, indicating that none of the studies were outliers or had a disproportionate impact on changes in the marginal bone level summary estimate (Table 5).

TABLE 5.

Leave‐one study‐out sensitivity analysis: Changes in marginal bone level.

Removed study	Overall MD (95% CI)	p	Heterogeneity
Lv et al. 2021	−0.03 (−0.13, 0.08)	0.60	Not applicable
Nastri et al. 2021	−0.01 (−0.14, 0.11)	0.81	Not applicable
Shi et al. 2020	−0.01 (−0.11, 0.09)	0.87	p = 0.98; I 2 = 0%
Yang et al. 2023	−0.05 (−0.14, −0.05)	0.33	p = 0.98; I 2 = 0%
			p = 0.98; I 2 = 0%

Abbreviations: CI: confidence interval, MD: mean difference.

4. Discussion

4.1. An Overview of the Findings

ASC‐retained and cement‐retained implant crowns were compared in this review based on changes in marginal bone level, periodontal parameters, aesthetic outcomes, technical complications, implant failure rate, and the necessity for augmentation at implant placement. ASC‐retained crowns were associated with minimal changes in marginal bone level, bleeding on probing, probing pocket depths, and pink and white aesthetic scores when compared to cement‐retained implant crowns, and the difference between the two groups was not statistically significant, with the exception of bleeding on probing. Although there were fewer instances of technical issues with cement‐retained implant crowns than ASC‐retained ones, the difference was neither statistically nor clinically significant. Neither intervention had an impact on the need for augmentation at implant placement or implant failure.

4.2. Quality of Evidence

Although different study designs, randomized and nonrandomized studies, were included in this systematic review, stringent selection criteria were applied to enhance the overall search quality and reduce heterogeneity. By doing so, the included studies were as comparable as possible, thereby increasing the reliability of our findings. In this context, the comparison was limited to studies that reported on single implant crowns in nonmolar sites, which helped to maintain a focused and relevant scope for our analysis. The quantitative analysis presented in this review was based on studies that ranged in quality from low to moderate. Despite this variation in quality, significant heterogeneity was only observed in the meta‐analysis of the pink aesthetic score, indicating acceptable homogeneity among the other included studies. The main reason for the observed homogeneity was likely related to the use of comparable methods of assessment to evaluate implant crown restorations of similar prosthetic designs in the aesthetic zone. This consistency in assessment methods contributed to the robustness of our findings.

Some concerns were expressed in one included randomized controlled trial [28], as the information on allocation concealment and blinding was not sufficient. This lack of detailed reporting raises questions about the potential for selection and performance biases, which could affect the validity of the study's findings. The remaining studies were nonrandomized, which inherently carries a higher risk of bias compared to randomized controlled trials. Amongst these, one study [48], was considered to have a low risk of bias in all assessed domains. This study demonstrated rigorous methodology, including clear reporting on participant selection, intervention implementation, and outcome assessment, enhancing its results' credibility. The remaining two studies [29, 45] did not adequately describe the blinding of data assessors. The absence of blinding in these studies introduced a high risk of detection bias, as the knowledge of intervention allocation could influence the assessment of outcomes. The lack of blinding, combined with other potential methodological weaknesses, led to these studies being rated at a high risk of bias. Such biases can significantly impact the reliability of the findings, making it crucial to interpret the results with caution.

Despite the small number of included studies, the precision of treatment effects has been improved by the use of similar methodologies in assessing changes in marginal bone level, periodontal parameters, and aesthetic outcomes. These methodologies include the use of periapical radiographs, CBCT, periodontal probes, and digital photography. By employing these standardized and reliable assessment tools, the accuracy and comparability of the data were enhanced, thereby strengthening the overall conclusions of our review.

4.3. Applicability of Evidence

The present systematic review has shown that the use of ASC‐retained implant crowns in nonmolar sites may contribute to the health of the peri‐implant tissue by reducing changes in bleeding on probing while achieving comparable outcomes to cement‐retained implant crowns in terms of probing pocket depth and aesthetics. The need for augmentation procedures at the time of implant placement was not influenced by the retention mechanism of the crowns (ASC vs. cement retention).

Limited marginal bone loss was observed over a follow‐up period between 12 and 44 months, with negligible difference between the two groups (0.03 mm). The levels of marginal bone changes were in accordance with other studies [33, 36, 37, 50] that also reported bone loss of 0.18–0.41 mm around implants restored with ASC restorations in the first year of loading. In the present review, a statistical significance was only evident for bleeding on probing, where less bleeding was noted in the ASC group. The difference can be attributed to the presence of residual cement in the cement‐retained group, which is likely to be associated with peri‐implant mucositis [14, 51]. Nevertheless, studies comparing screw‐ to cement‐retained implant‐supported prostheses over a 3‐year duration [52, 53] did not show significant differences in bleeding on probing or the rate of peri‐implant mucositis.

The aesthetic outcomes of implant prostheses are more influenced by the width of keratinized tissue [54], buccal bone thickness [55] and implant crown form and color [56] than the retention system used. The fact that there was no statistically significant difference between the two groups is therefore not surprising. Our findings were consistent with other studies [57, 58] that recorded pink and white aesthetic scores of screw‐ and cement‐retained implant‐supported prostheses in the anterior maxilla. Although the use of ASC was not part of the implant design, no significant findings were observed between the two retention systems. Compared to cemented crowns, implant crowns with ASC were linked to higher rates of screw loosening and porcelain fracture. Even though the difference was not statistically significant, these findings might highlight important clinical considerations. The increased incidence of porcelain fracture in ASC crowns can be attributed to the thinner porcelain required by the design of ASC [30, 32]. This design necessitates a compromise in material thickness to accommodate the ASC, potentially making the porcelain more susceptible to fractures under occlusal forces. The rate of porcelain fracture observed in the included studies aligns with several other studies [31, 37, 38], suggesting a consistent pattern across different research settings. However, it is noteworthy that the fracture rate reported in these studies was lower than that recently published on ASC [36]. The relatively higher rate of porcelain fracture in the latter report was attributed to the noncompliance of one bruxer with night guard wear that resulted in six crown fractures.

Several systematic reviews compared the clinical outcomes of screw‐retained to cement‐retained implant prostheses [13, 14, 16]. Their findings showed similar rates of survival, complications, and maintenance issues. While screw‐retained prostheses were more prone to technical complications, cement‐retained prostheses had more biological complications. Overall, screw‐retained prostheses were the preferred option mainly due to their retrievability despite the technical challenges encountered in the aesthetic zone. A notable short communication published by Strauss et al. [59] showed that the difference between cemented and screw‐retained implant crowns in terms of marginal bone loss was clinically negligible at 5 years of loading. In contrast to the current review's findings, the authors observed a minor marginal bone loss in the screw‐retained group and slight bone gain in the cemented group. They attributed this to the retrievability of the screw‐retained implant crown and their frequent removal and reinsertion over time in comparison with cemented ones. It is possible that the present review's limited follow‐up period and the small number of included studies were not enough to capture that finding.

In addition, two systematic reviews [26, 39] evaluated the clinical outcomes of ASC‐ and straight screw channel (SSC)‐retained implant prostheses. There was no significant difference in the occurrence of peri‐implantitis and other biological complications between ASC and SSC prostheses. ASC‐retained implant prostheses exhibited a higher incidence of screw loosening and fracture compared to SSC‐retained implant prostheses, but the difference was not statistically significant. The findings of both reviews suggest that while ASC‐retained implant prostheses offer flexibility in implant placement, clinicians should be aware of the potential for increased technical complications. In the present review, ASC‐retained implant crowns in single nonmolar sites were associated with lower changes in marginal bone levels, probing pocket depth, bleeding on probing, and aesthetic outcomes, though the differences were only significant for bleeding on probing. Therefore, the use of ASC‐retained implant crowns in nonmolar sites seems to be a reliable approach to maintain the retrievability advantage of the screw‐retained prosthesis without compromising peri‐implant health or the aesthetic outcome.

It is important to note that the present systematic review had inherent limitations, and its findings should be interpreted with caution. Firstly, the inclusion of a few studies as the existing body of research on ASC‐retained implant crowns is still limited and falls short in several areas that are clinically significant to our daily practice, particularly the availability of long‐term data on biological and technical complications. Secondly, including both randomized and nonrandomized controlled trials has improved the comprehensive search for evidence. However, the inclusion of different study designs might have diluted the results and reduced the overall quality, as nonrandomized studies are more prone to bias. Thirdly, the quality of the included studies varied significantly, which could impact the overall reliability. Fourthly, the observed minimal heterogeneity might have resulted from the small number of included studies rather than a true lack of heterogeneity. In addition, heterogeneity, albeit detected in one outcome, remains a concern due to variations in populations and methods of assessment. For example: the included studies might have employed various imaging modalities in measuring marginal bone levels, which might have resulted in discrepancies in reported changes in marginal bone levels and the interpretation of the data. Nevertheless, the included studies used similar calibration standards and measurement protocols, particularly in terms of using image analysis software.

Therefore, there is a critical need for more long‐term and robust randomized controlled trials to strengthen the evidence base and provide more definitive findings. These future studies should aim to standardize methodologies, include larger and more diverse samples, and extend follow‐up periods to better assess the long‐term outcomes and potential complications.

5. Conclusions

ASC‐retained implant crowns in nonmolar sites, within the limitations of this review, exhibit similar short‐term clinical, radiological, and aesthetic results to cement‐retained implant crowns, with the expected reduction in bleeding on probing. The results of this review need to be supported by further data from long‐term, carefully planned randomized controlled trials that follow CONSORT standards.

Author Contributions

Momen A. Atieh: concept/design, data collection, data analysis/interpretation, drafting article, critical revision of article, approval of article. Maanas Shah: data analysis/interpretation, critical revision of article, approval of article. Abeer Hakam: critical revision of article, approval of article. Asma Albalushi: data collection, data analysis/interpretation, drafting article, critical revision of article, approval of article. Anas Abdulmunim: data analysis/interpretation, critical revision of article, approval of article. Fawaghi AlAli and Nabeel H. M. Alsabeeha: critical revision of article, approval of article.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.