Table 3 IOS accuracy compared to conventional techniques

Lin et al. [64]

Partially dentate mandible with 2 implants and 4 different angulations (0,15,30 and 45 degrees)

Distance and angulation

In vitro (n = 4)

iTero

PVS (Aquasil) open tray, non-splinted

Cagenix

The amount of divergence between implants significantly affected the accuracy of the milled casts created digitally. The digital technique was more accurate when the implants diverged more. At 0 and 15 degrees of divergence, the digital method resulted in highly significantly less accurate definitive casts. At 30 and 45 degrees of divergence, however, the milled casts showed either no difference or marginal differences with casts created conventionally

Mangano et al. [44]

Partially edentulous maxilla with 3 implants

Full edentulous maxilla with 6 implants

3D surface

In vitro (n = 2)

Trios 2 (71 μm)

CS 3500 (47-63 μm)

Zfx Intrascan 117-103 μm)

Planscan (233-253 μm)

N/A

Iscan D104I (Imetric3D)

CS3500 most accurate IOS but no SS compared to TRIOS

Refractory Index of PEEK is better than Titanium

Flügge et al. [43]

Partially dentate mandible with 2 implants

Partially dentate mandible with 5 implants (Kennedy 1)

Distance and angulation

In vitro (n = 2)

Trios

iTero

True Def

N/A

D250

The precision of IOS decreases with longer distances between scanbodies, especially crossing the midline

Fukazawa et al. [45]

Partially dentate mandible with 2 implants (short and long distance)

Distance

In vitro (n = 2)

Trios 2 (7 and 20 μm)

Lava COS (27 μm and 80 μm)

True Def (17 μm and 60 μm)

Kavo ARCTICA (3 μm and 18 μm)

N/A

CMM

UPMC 550-Sarat

Trios comparably accurate to Lab scanner and SS more accurate than the other IOS tested

For longer distances, IOS accuracy decreases

Deviation of up to 100 μm is acceptable

Basaki et al. [21]

Partially dentate mandible with 4 implants (Kennedy 1)

Distance

In vitro (n = 1)

iTero (116 μm)

PVS monophase with custom trays (56 μm)

D810

PVS impressions were statistically more accurate than IOS

Implant angulation did not affect IOS accuracy. Milled 3D casts were less accurate compared to stone casts

Imburgia et al. [46]

Partial maxillary arch with 3 implants

3D surface

In vitro (n = 1)

Trios 3 (Tr = 50 μm/Pr = 24 μm)

CS3600 (Tr = 45 μm/Pr = 24 μm)

Omnicam (Tr = 58 μm/Pr = 26 μm)

TrueDef (Tr = 61 μm/Pr = 19 μm

N/A

ScanRider

CS3600 had SS higher trueness compared to other IOS. No SS differences in precision were found

Accuracy in the partial arch is higher for all IOS compared to the Full arch situation

Chew et al. [24]

Partial jaw with 2 implants and 2 different depths

Distance and angulation

In vitro (n = 2)

Trios

True Def

iTero

Polyether mono (custom tray)

CMM

Model Global Silver

Conventional impressions had ss less deviation compared to IOS. Implant depth affected IOS accuracy. Angulation did not affect accuracy

Chia et al. [25]

Partial jaw with 2 implants and 3 different angulations

Distance and angulation

In vitro (n = 3)

Trios (31-45 μm) depending on configuration

Polyether mono (custom tray) (18-33 μm) depending on configuration

CMM

Model Global Silver

Distortions were found with conventional and IOS imps. Conventional imps in parallel implants had highest accuracy compared to IOS. Angulation affects IOS accuracy

Marghalani et al. [31]

Partially dentate mandibles with 2 implants

3D surface

In vitro (n = 2)

Omnicam (33-55 μm)

True Def (27-39 μm)

Polyether mono on splinted implant copings (open tray) (26-53 μm)

Activity 880 industrial scanner

True Def IOS was more accurate but SS difference were not always observed

Low deviations < 56 μm

Kim et al. [47]

Partially dentate mandible with 6 implant cylinders

Distance

In vitro (n = 1)

Trios 3

Omnicam

CS3600

I500

iTero Element

N/A

StereoSCANneo

All IOSs exhibit deviations as scanning distance increases from the start position

Trios3 and Medit outperformed other IOSs for partially edentulous accuracy

Mangano et al. [23]

Partial edentulous maxilla with 2 implants

3D Surface

In vitro (n = 1)

Trios 3 (Tr = 28 μm/Pr = 21 μm)

CS3600 (Tr = 23 μm/Pr = 17 μm)

Omnicam (Tr = 38 μm/Pr = 43 μm)

DWIO (Tr = 49 μm/Pr = 34 μm

Emerald (Tr = 49 μm/Pr = 29 μm)

N/A

Freedom DOF

Trios3 and CS3600 were SS more accurate compared to other IOS

Accuracy of IOS in implants complete arch is NOT corelated to IOS resolution

Motel et al. [65]

Titanium partial model with 3 implants and 3 different scanbody designs and 2 different scan strategies

Distance and 3D surface

In vitro (n = 1)

Trios 3

N/A

ATOS So4 II

All in One scan strategy produced more accurate results (71 μm)

Emergence profile scan produced lower accuracy (125 μm)

In All in One scan strategy, cylindrical scanbodies with flatter surfaces produced more accurate results

Alsharbaty et al. [32]

Partially dentate mandibles and maxillae with 2 posterior adjacent implants

Distance

In vivo (n = 28)

Trios 3

PVS (Panasil) dual mix, plastic tray/splinted (used as reference)

PVS (Panasil) dual mix, plastic tray/non splinted in open and closed tray methods (used for comparison)

CMM (Dea Global)

Conventional open tray pick-up impression was ss more accurate compared to IOS and conventional closed tray pick-up impression

Jiang et al. [92]

Partial dentate maxilla and mandible with implants and 2–4 teeth span

3D surface

In vivo (n = 31)

Trios (27 μm)

Material not provided/splinted, open tray

D800

The 3D discrepancy between digital and traditional impression is within clinical acceptable range