conventional drills vs piezoelectric surgery preparation

147

Eur J Oral Implantol 2017;10(2):147–158

RANDOMISED CONTROLLED CLINICAL TRIAL

Marco Esposito, DDS, PhDFreelance researcher and Associate Professor, Depart-ment of Biomaterials, The Sahlgrenska Academy at Göteborg University, Sweden

Carlo Barausse, DDSResident, Department of Biomedical and Neuromotor Sciences, Unit of Periodon-tology and Implantology, University of Bologna, Bolo-gna, Italy

Andrea Balercia, DDSResident in Maxillofacial Surgery, Azienda Ospedaliera Ospedali Riuniti di Ancona, Ancona, Italy

Roberto Pistilli, MDResident, Oral and Maxil-lofacial Unit, San Camillo Hospital, Rome, Italy

Daniela Rita Ippolito, DDSPostgraduate Student, Department of Orthodontics, School of Dentistry, Univer-sity of Brescia, Brescia, Italy

Pietro Felice, MD, DDS, PhD Researcher, Department of Biomedical and Neuromotor Sciences, Unit of Periodon-tology and Implantology, University of Bologna, Bolo-gna, Italy

Correspondence to: Dr Marco Esposito, Casella Postale 34, 20862 Arcore (MB), Italy. E-mail: [email protected]

Marco Esposito, Carlo Barausse, Andrea Balercia, Roberto Pistilli, Daniela Rita Ippolito, Pietro Felice

Conventional drills vs piezoelectric surgery preparation for placement of four immediately loaded zygomatic oncology implants in edentulous maxillae: results from 1-year split-mouth randomised controlled trial

Key words atrophic maxilla, implant site preparation, piezoelectric surgery, zygomatic implants

Purpose: To compare the outcome of site preparation for zygomatic oncology implants using con-ventional preparation with rotary drills or piezoelectric surgery with dedicated inserts for placing two zygomatic implants per zygoma according to a split-mouth design. Materials and Methods: Twenty edentulous patients with severely atrophic maxillas not having suf-ficient bone volume for placing dental implants and less than 4 mm of bone height subantrally had their hemi-maxillas randomised according to a split-mouth design into implant site preparation with conventional rotational drills or piezoelectric surgery. Two zygomatic oncology implants (unthreaded coronal portion) were placed in each hemi-maxilla. Implants that achieved an insertion torque super-ior to 40 Ncm were immediately loaded with screw-retained metal reinforced acrylic provisional pros-theses. Outcome measures were: prosthesis and implant failures, any complications, time to place the implants, presence of post-operative haematoma, and patient’s preference by independent assessors. All patients were followed up to 1 year after loading. Results: In two patients drills had also to be used at the piezoelectric surgery side to enable implant sites to be prepared. One implant for the conventional drill group did not achieve an insertion torque super-ior to 40 Ncm since it fractured the zygoma. No patients dropped out and two distal oncology implants failed in the same patient (one per group), who was not prosthetically rehabilitated. Six complications occurred at drilled sites and three at piezoelectric surgery sites (two patients had bilateral complica-tions), the difference being not statistically significant (P (McNemar’s test) = 0.375; odds ratio = 4.00; 95% CI of odds ratio: 0.45 to 35.79). Implant placement with convention drills took on average 14.35 ± 1.76 min and with piezoelectric surgery 23.50 ± 2.26 min, implant placement time being sig-nificantly shorter with conventional drilling (difference = 9.15 ± 1.69 min; 95%CI: 8.36 to 9.94 min; P < 0.001). Post-operative haematomas were more frequent at drilled sites (P = 0.001), and 16 patients found both techniques equally acceptable, while four preferred piezoelectric surgery (P = 0.125).Conclusions: Both drilling techniques achieved similar clinical results, but conventional drilling required 9 min less and could be used in all instances, although it was more aggressive. These results may be system-dependent, therefore they cannot be generalised to other zygomatic systems with confidence.

Conflict-of-interest statement: This study was partially supported by Southern Implants (Irene, South Africa), the manufacturer of the zygomatic implants and the conventional drills evaluated in this study. However, data property belonged to the authors and by no means did the manufacturer interfere with the conduct of the trial or the publication of its results. Drs Felice and Pistilli devel-oped the piezoelectric surgery zygomatic insert used in the present study.

Esposito et al Drills vs piezoelectric surgery for zygomatic implants148


the tip. It was therefore interesting to compare the clinical outcome of zygomatic implants placed using either the new purposely developed piezoelectric surgery inserts or the conventional drills (see Fig 2) to see whether one of the two preparation proced-ures was preferable.

The aim of this randomised controlled trial (RCT) of split-mouth design was to compare the clinical outcome of site preparation for zygomatic oncol-ogy implants using conventional preparation with rotary drills or piezoelectric surgery for placing two zygomatic implants per zygoma. It was planned to report data 5 years after loading and this is the first of the planned publications. This article is reported according the CONSORT statement for improving the quality of reports of parallel-group randomised trials (http://www.consort-statement.org/).

Materials and methods

Trial design

This was a two-centre RCT of split-mouth design with balanced randomisation and independent assessment. The hemi-maxilla of each patient was randomly allocated to have two adjacent zygo-matic oncology implant sites prepared either with the two piezoelectric surgery inserts (Zygoma kit, Esacrom, Imola, Italy) (piezo group; Fig 1a and b), or with the conventional drills (Fig 2) of the zygomatic implant manufacturer (Southern Implants, South Africa)(drill group).

Eligibility criteria for participants

Any patient with a severely atrophic edentu-lous maxilla not allowing the placement of dental implants and having a residual bone height under the maxillary sinus less than 4 mm (Fig 3a and b) as measured on cone-beam computer tomography (CBCT) or conventional computer tomography (CT) scans requesting a fixed prosthesis who was 18 or older and able to understand and sign an informed consent form, was eligible for inclusion in the trial. Coronal slices were added to conventional CBCT/CT scans to evaluate the ostiomeatal complex and the sinus epithelium conditions. Only patients with

Introduction

Zygomatic implants are an alternative to conven-tional bone augmentation and implant rehabilitation for severely atrophic maxillas1. Zygomatic implants are long implants which are placed in the zygo-matic bone passing or not through the maxillary sinus depending on the individual local anatomic conditions2. Oncology implants are a variation of zygomatic implants used in various maxillofacial reconstructions procedures and characterised by an unthreaded coronal portion of various lengths. Two zygomatic implants can be placed in each zygoma and can be successfully immediately loaded if inserted with a sufficient torque3. This is a major potential advantage over conventional bone aug-mentation procedures used to rehabilitate severely atrophic maxillae since patents could be function-ally rehabilitated in a single day instead of undergo-ing two to three surgical procedures over several months4,5. Despite zygomatic implants having been in use for almost 20 years1,6-9, reliable comparative trials evaluating the actual efficacy and potential risks compared with conventional augmentation proced-ures are still lacking10. While running a randomised controlled trial (RCT) comparing four immediately loaded zygomatic implants with conventional implants placed in bone augmented with bone sub-stitutes (data under evaluation), it was apparent that one of the problems was the difficulty in control-ling the drilling procedures for zygomatic implants using such long drills, due to the risk of waving of the drill (personal communications). The authors felt that the drilling precision could be improved and developed two specific tips to be used with a piezo-electric surgery device (see Fig 1a and b). The two different piezoelectric inserts have double internal irrigation and total lengths of 74 or 94 mm with a first angled part and a straight part. The two inserts differ in terms of tip shape: conical (2.9 mm diam-eter; Fig 1a) with micro-sharp blades and diamond-shape (3.5 mm diameter; Fig 1b) with blades. These zygomatic inserts are manufactured from medical stainless steel with T-black finishing, obtained with a double-coated nanostructure surface treatment that allows heating tissue reduction and has greater resistance to wear and corrosion. Both inserts have depth indicators, at 30, 35, 40, 45, and 50 mm from

Esposito et al Drills vs piezoelectric surgery for zygomatic implants 149


healthy sinuses were asked to join the trial. Patients were not admitted to the study if any of the follow-ing exclusion criteria was present: • General contraindications to implant surgery;• Irradiated in the head and neck region with more

than 70 Gray;• Immunosuppressed or immunocompromised

patients;• Treated or under treatment with intravenous

amino-bisphosphonates;• Untreated periodontal disease;• Poor oral hygiene and motivation;• Uncontrolled diabetes;• Pregnant or lactating;• Substance abusers;• Psychiatric problems;• Lack of opposite occluding dentition/prosthesis;• Restricted mouth opening (less than 3.5 cm inter-

arch anteriorly);• Acute or chronic infection/inflammation in the

area intended for implant placement;• Unable to commit to a 5-year follow-up;• Patients participating in other studies, if the pre-

sent protocol could not be properly adhered to;• Referred only for implant placement.

Patients were categorised into three groups accord-ing to what they declared: non-smoker, moderate smoker (up to 10 cigarettes per day) and heavy smoker (more than 10 cigarettes per day).

Setting and locations

Patients were seen at the University Hospital of Ancona, Italy, by Dr Pietro Felice and Dr Roberto Pistilli, who each treated 10 patients and followed similar procedures.

Surgical procedures

All patients received thorough explanations, under-stood the nature of the study and signed a written informed consent form prior to be enrolled in the trial. Stereolithographic models of the maxillae were created from CBCT/CT scans to better plan the im-plant insertion angles (Fig 3c). Anatomical landmarks to be avoided, such as the infraorbital foramens and the correct implant insertion axes, were marked with

Fig 1a and b Inserts for piezoelectric surgery used in this study: a) the conical insert with micro-sharp blades; b) the diamond-shape insert.

Fig 2a to d The set of drills used in this study in order of sequence of use: a) zygomatic round ball for initial prep-aration; b) 2.9 mm diameter pilot drill; c) 3.4 mm diameter countersink drill; d) 3.4 mm diameter drill.

a b

ba

dc



a pencil. Diagnostic wax-up and surgical guides were prepared to help clinicians select the most appropri-ate position and angulation of each implant. Efforts were made to plan implant exits at crestal level rather than palatally.

Patients rinsed with 0.2% chlorhexidine mouth-wash for 1 min prior to implant placement. They were treated under general anaesthesia and received intravenous prophylactic antibiotic therapy (1 g of amoxicillin plus clavulanic acid, or, if allergic to peni-cillin, 300 mg clindamycin). Articaine with adrenaline 1:100.000 was injected locally to reduce bleeding and increase visibility. The right side of the patient was treated first. After crestal and release incisions, a mucoperiosteal flap was elevated, the maxilla was exposed to allow the identification of the infraorbital foramen and of the incisura between the zygomatic arch and the lateral and medial surface of the fron-tal process of the zygomatic bone. If necessary, a 10 × 5 mm or wider window, extending from the sinus floor to the base of the zygomatic bone, was opened on the lateral wall of the maxillary sinus close to the infra-zygomatic crest. The sinus lining was carefully lifted and an absorbable haemostatic gela-tine sponge (Spongostan Special, Ethicon, Johnson & Johnson, New Brunswick, NJ, USA) was used to protect the sinus epithelium after its displacement. This sponge, of porcine origin, is sterile, water-insol-uble and malleable, intended for haemostatic use by applying to a bleeding surface. The sponge is off-white and porous in appearance. At this point, the right side of the patient was randomly allocated to either conventional rotary drill (Fig 3d and e) or piezoelectric surgery preparation (Fig 3f) following the indications contained in the sequentially num-bered envelope corresponding to the recruitment number of the patient. The implant site preparation of the left side was accomplished with the other procedure. Each patient received four zygomatic oncology implants only, two per side. Those sites randomised to rotary drills were prepared following the instructions of the zygomatic implant manufac-turer (Southern Implants). The zygomatic round ball was used for the initial preparation, followed by the 2.9 mm diameter pilot drill to full depth (penetration of the outer cortical layer of the zygomatic bone at the zygomatic malar face). The length of the implant used was determined with a depth indicator. Then

the 3.4 mm diameter countersink drill was used, fol-lowed by a 3.4 mm diameter drill to full depth (see Fig 2).

The two sites randomly allocated to piezoelectric surgery preparation were initially prepared with the conical-shape tip insert (Fig 1a) which was inserted, when possible, in the residual alveolar bone exiting at the base of the sinus window. In this way, implants had an endosseous path at the level of the atrophic crest, which provided better primary stability. Due to its length, the tip of this insert was subsequently passed through the maxillary sinus, entering the mid-portion of the zygomatic body and emerging from the zygomatic malar face. The diamond-shape tip insert (Fig 1b) was then used to widen the implant site to the desired diameter (3.5 mm) in relation to the implant diameter (4.3 mm). A depth indicator was inserted into the site to determine the correct length of the zygomatic implants.

The zygomatic implants used were of the oncol-ogy type, characterised by lack of threads in the coronal portion for 7.5, 12.5 17.5, 22.5, 27.5, 30.0, 32.5 and 35.0 mm, depending on the implant length, and a 55° angled neck with external hexagon connection (Fig 3 g). The following implant lengths were used: 27.5, 32.5, 37.5, 42.5, 47.5, 50.0, 52.5 and 55.0 mm, all having a diameter of 4.3 mm.

When anatomical conditions allowed, ideally implants were not inserted through the sinus cavity, but into or on to the bone externally to the sinus. Surgical templates were used to position the implant exit into the oral cavity at crestal level and not on the palate. A retractor was placed to the incisura between the zygomatic arch and the lateral and medial sur-face of the frontal process of the zygomatic bone to facilitate the correct three-dimensional orientation of the implant. While preparing the implant sites, abundant irrigation was provided. After the first im-plant was placed, the same procedure was repeated to place the second implant. It was attempted to position the implant apexes at least 3 mm apart and to place implants with an insertion torque superior to 40 Ncm in order to be able to load them immedi-ately. Implants inserted with less than 40 Ncm were submerged for 5 months. Bicortical engagement was always obtained meaning that the tip of the implant protruded between 1 and 2 mm on the other side of the zygoma. The Bichat’s fat pads were exposed and



Fig 3a to r a and b) details of one preoperative CT scan of one of the patients included in the study showing the subantral bone height at right and left sides, respectively inferior to 4 mm; c) stereolithographic model of the maxilla used to plan implant insertion; d) right maxilla randomly allocated to the drill group; e) the first oncology zygomatic implant in site; f) preparation with piezo-surgery; g) the zygomatic oncology implant characterised by lack of threading in the coronal portion; h) the four zygomatic oncology implants in place; i) the Bichat’s fat pads were mobilised to cover the implants; j) flaps were sutured around the impression copings; k) a brass wire was used together with self-curing acrylic resin to stabilise the impression copings; l) panoramic radiograph, m) frontal and n) occlusal pictures taken at initial loading with the provisional prosthesis; o) pic-tures taken at 3 days post-surgery to compare the extension of the haematomas: the drilled side showed a larger haematoma; p) panoramic radiograph; q) implant-supported prosthesis; and r) face of a patient 1-year after loading.

ba c

ed f

hg i

kj l

m n

p

o

q

r



gently shifted medially to cover the exposed implant portions (Fig 3 h to i). Flaps were then sutured with simple interrupted 4-0 resorbable sutures (Vicryl, Ethicon FS-2, Johnson & Johnson, USA) around the impression copings (Fig 3j).

Prosthetic procedures

Prosthetic procedures were initiated immediately after flap suturing. Panoramic radiographs were taken to verify proper seating of all the impression copings. A self-curing acrylic resin (Pattern Resin, GC America, Alsip, IL, USA) was positioned on a brass wire, to further stabilise the impression cop-ings in position (Fig 3k). A pick-up impression was taken using a polyether material (Impregum, 3M ESPE, Milan, Italy), and when possible the patient denture, with holes in the resin palate was used as a customised tray. Healing caps were positioned. A screw-retained metal reinforced acrylic cross-arch provisional prosthesis was delivered within 1 week (Fig 3l to n). The following post-surgical instructions were given:• Augmentin 1 g (or Clindamycin 300 mg) to be

taken three times a day for 1 week.• Ibuprofen 600 mg was prescribed to be taken

four times a day during meals for 1 week, but patients were instructed not to take them in the absence of pain.

• Xylometazolin hydrochloride (nasal decongest-ant) 1 mg. Five drops to be taken three times a day for 2 weeks;

• A soft diet was recommended for 2 weeks.• 0.2% chlorhexidine gentle rinses twice a day for

2 weeks.

Patients were recalled and checked after 3 days (also to evaluate haematoma and patient preference; see Fig 3o); after 10 days (for suture removal and pros-thesis check-up), and after 1 month.

Five months after delivery, the provisional pros-theses were to be removed, implant stability was to be checked by tightening the abutment screws with a 15 Ncm torque using a manual torque wrench, and a definitive impression at abutment level was to be taken using a rigid impression material and impres-sion copings with an open tray. Within 1 month definitive screw-retained metal-resin/composite

definitive cross-arch prostheses were to be delivered. However, due to financial problems not a single de-finitive prosthesis was delivered up until the first year in function (Fig 3p to r).

Patients were enrolled on an oral hygiene pro-gramme, with recall visits every 6 months. Opera-tors were free to increase maintenance frequency (every 3 to 4 months) based on individual needs. Dental occlusion was evaluated at each maintenance visit. An independent outcome assessor (initially Dr Lorenzo Tuci, but replaced by Dr Carlo Barausse) conducted follow-ups.

Outcome measures

Outcome measures were:• Prosthesis failure defined as no prosthesis deliv-

ery or prosthesis replacement because of implant failures. It was indicated whether the loss of the prosthesis was due to implant failures according to the implant site preparation procedures.

• Implant failure defined as an implant displaying rotational mobility, any infection dictating im-plant removal, and/or any mechanical complica-tion rendering the implant unusable (e.g. implant fracture or deformation of the connecting plat-form). Implant stability assessments were to be done, with the removed prostheses, at a delivery of the definitive prosthesis (this procedure actu-ally did not take place as originally planned, since no definitive prosthesis was delivered) and 1 year after loading by tightening the abutment screws with a 15 Ncm torque. Rotating implants were considered failures and were to be removed. It was possible that a few zygomatic implants dis-played a slight horizontal mobility due to their lengths and possible lack of alveolar bone at their exits. This was recorded, but the implants, if not rotating, were considered to be successful.

• Biological or prosthetic complications. Compli-cations were divided between those possibly caused by the implant site preparation and those possibly independent from the implant site prep-aration.

• Time for site preparation defined as the time ne-cessary to prepare and fully insert both implants at each site was recorded, starting from the first drill to complete implant insertion.



• Presence of haematoma assessed on frontal pic-tures of the patient’s face taken at day 3 post-surgery by a blinded outcome assessor. The presence and extension of the haematoma at each site of the face was evaluated to see if one of the two sites had a larger haematoma. Hae-matomas were scored in the following manner: 1) larger haematoma at the rotary drill treated side; 2) larger haematoma at the piezoelectric surgery treated side; 3) no haematoma present at both sides; 4) both haematomas looking similar in extension.

• Patient preference was also evaluated at day 3 post-surgery by the independent outcome asses-sor, who asked patients if they preferred one of the two sides, and for what reason. The follow-ing question was asked: which side of the max-illa did you prefer? Patients pointed to the pre-ferred site and the outcome assessor recorded the patient’s choice, which could have been: 1) the side prepared with piezoelectric surgery; 2) the side prepared with traditional drills; 3) none, both treatments were equally good; 4) none, both treat-ments were equally bad. Patients’ explanations/comments on the procedures were recorded.

Sample size, random sequence and allocation concealment

No sample size was calculated and our recruitment capacity was limited to 20 patients.

Two computer-generated restricted randomi-sation lists were created. A blocked randomisation was applied to include 10 patients at each centre. Only one of the investigators (Dr Esposito), who was not involved in the selection and treatment of the patients, was aware of the randomisation sequence and could have access to the randomisation list stored in his password-protected portable computer. The randomised codes were enclosed in sequentially numbered, identical, opaque, sealed envelopes. Envelopes were opened sequentially after flap ele-vation; meaning treatment allocation was concealed to the investigators in charge of enrolling and treat-ing the patients.

One clinician, (initially Dr Tuci, but replaced at the end of the treatment phase by Dr Barausse), who were not involved in the treatment of the patients,

measured implantation time, implant stability and patient preference. Complications were registered and treated by surgeons in a non-blinded mode. Dr Barausse made all evaluations on haematomas on pictures without knowing group allocation. Since patients were under general anaesthesia they were blinded to treatment allocation.

Statistical methods

All data analysis was carried out according to a pre-established analysis plan. A dentist (Dr Ippolito) with expertise in statistics analysed the data, without knowing the group codes. The patient was the stat-istical unit of the analysis. An intention-to-treat ana-lysis was applied, but an as-treated analysis (carried out removing the data of the two patients who had both sites treated with conventional drills) was also done. McNemar’s test and Fisher’s exact test were used to investigate the differences in the proportion of patients with prosthesis failures, implant failures and complications (dichotomous outcomes) respect-ively between the implant site preparation methods and between the clinicians. Differences at patient level in the time to prepare the implant site (con-tinuous outcome) were investigated through paired t tests (comparisons between implant site preparation methods) and Mann-Whitney U tests (comparisons between the clinicians). McNemar’s test was used to investigate differences in the implant preparation site preference and haematoma between the implant site preparation methods. All statistical comparisons were conducted at a 0.05 level of significance.

Results

Twenty-five patients were screened for eligibility, but five patients were not enrolled in the trial: three patients had more than 4 mm of bone subantrally and two had maxillary sinus pathologies. Twenty patients were considered eligible and were consecu-tively enrolled and treated on the trial. All patients were treated according to the allocated interven-tions, with the exception of two patients who had both sites prepared using conventional drills, because the first piezo surgery insert was unable to penetrate the bone (in one case the preparation



was completed with drills only, while in the second case it was completed with piezoelectric surgery). No patient dropped out and data of all patients was evaluated in the statistical analyses. The following additional protocol deviations occurred: • No patient received the definitive prosthesis dur-

ing the first year in function because of financial problems.

• Two patients had both sites prepared with conventional drills as well since the piezoelec-tric surgery insert was unable to perforate the bone.

• One of the 47.5 mm long implants in the piezo-group was too long, so its apex was cut by 2.5 mm at surgery with laboratory disks (see Fig 4).

Patients received zygomatic implants from May to December 2015. The follow-up of all patients was up to 1 year after prosthetic loading.

There were 11 females and nine males in the study. The mean age at implantation was 68 (range 62 to 75). Fourteen patients were non-smokers and six were moderate smokers. The main base-line patient characteristics are presented in Table 1. There was no apparent systematic baseline imbal-ance between the two groups. Only one implant in the conventional drill group did not achieve an inser-tion torque superior to 40 Ncm, because it fractured the zygoma (Fig 5) and was left unloaded.• Prosthetic failures. One provisional prosthesis was

lost because both distal zygomatic implants failed 15 days after their placement. Two 6 × 4 mm implants were placed at lateral incisive position, but no new implant-supported prosthesis was delivered as in the meantime the patient was diagnosed with prostate cancer.

• Implant failures. The two distal zygomatic implants (one per group) failed in the same patient 2 weeks

Table 1 Intervention characteristics (20 patients).

Rotational drill group Piezoelectric surgery group

Total number of inserted implants 20 20*

Number of 27.5 mm long implants 1 0








Implants inserted with a torque superior to 40 Ncm 39 40

Implants inserted with a torque up to 40 Ncm 1 0

*Four implants were inserted after drill preparation, as the piezoelectric surgery insert was unable to perforate the bone.

Fig 4 Immediate post-loading panoramic view of the patient who had the mesial left implant cut before insertion because the correct implant length was not in stock.

Fig 5 Clinical view of the zygoma fractured at placement of the implant (drill group). An insertion torque superior to 40 Ncm was not obtained; therefore the implant was sub-merged whereas the remaining three implants were loaded immediately. It was planned to load the submerged implant at delivery of the definitive prosthesis, which has not yet been delivered.



after their placement. The patient had swelling, pain and infection signs at both implant sites and both implants were mobile and removed. Two 6 x 4 mm Southern implants were placed in tooth positions 12 and 22. The patient also complained about discomfort at the two mesial zygomatic implants, which were stable. Antibiotic therapy (amoxicillin, plus clavulanic acid 1 g three times a day for 10 days) was prescribed. While waiting for the healing period a prostatic carcinoma was diagnosed and the patient suspended the oral rehabilitative procedures.

• Complications. Six complications occurred at drilled sites and three at piezoelectric surgery sites (two patients had bilateral complications), the difference being not statistically significant (P (McNemar’s test) = 0.375; odds ratio = 4.00; 95% CI of odds ratio: 0.45 to 35.79).

The complications at the drilled sites were:• Two patients had burns to the inferior lips

(Fig 6a to c) because of the drills, which healed spontaneously.

• One patient had a fracture line at the zygoma at placement of implant in position 15 (Fig 5). The desired torque could not be obtained and the implant was submerged and was still sleeping at the 1-year follow-up.

• One patient had infection signs (pain and swell-ing) at both distal implants, which were mobile 15 days after their placement. Both implants were removed as previously described.

• Two patients had a burning sensation and discom-fort at implants in position 26 at 6 months after loading; the mucosa was inflamed and bleeding (mucositis). The prostheses were removed and cleaned together with the implant abutments. Patients were instructed to use chlorhexidine gel instead of toothpaste for 3 weeks, three times a day and the problem was solved.

The complications at the piezoelectric surgery sites were: • One patient had infection signs (pain and swell-

ing) at both distal implants, which were mobile 15 days after their placement. Both implants were removed as previously described.

• One patient showed an implant fenestration at the implant in position 25 one month after its placement (see Fig 7). The fenestration is still present, stable and inflammation free. No treat-ment was attempted.

• One patient reported discomfort and being able to feel the implant tip, extraorally, at the left zygoma level 1-and-a-half months after load-ing. She also complained she could not sleep on

Fig 6a to c a) preoperative view; b) view at 3 days post-surgery, showing the burn mainly at the lower lip at the drill- treated side; c) 1-year view, no signs of the burn are visible.

Fig 7 Implant fenestration (piezo group) at 1 month after placement. No pathologies of the surrounding soft tissues can be observed. No augmentation procedure was attempted to close the fenestration. The fenestration remained stable over the first year.

cba



that side and that a solution had to be found. The tip of implant 25 could be felt extraorally, all implants were stable and symptom-free. It was decided to cut off the implant tip by opening a flap under the lower eyelid as a blepharoplasty to visualise the area of interest (Fig 8a to c). The im-plant apex extruded outside the zygoma by four threads. The protruding portion of the implant was removed with a diamond ball bur and the area was then cleaned with hydroxide peroxide to remove the residuals of titanium powder. The patient healed well and had no more discomfort.

• Mean implant insertion time: on average it took 23.50 ± 2.26 min to prepare and place implants at piezo-surgical sites and 14.35 ± 1.76 min at con-ventionally drilled sites, the difference being stat-istically significant (difference = 9.15 ± 1.69 min; 95%CI: 8.36 to 9.94 min; P < 0.001). By remov-ing the two patients who had both sites treated or partially treated with drills, the difference was still significant (difference = 9.11 ± 1.78 min; 95%CI: 8.23 to 10.00 min; P < 0.001).

• Haematoma at day 3: Post-operative haematomas were larger at drilled sites in 11 patients and similar

at both sides in nine patients. Significantly more haematomas occurred at conventional drilled sites (P = 0.001). By removing the two patients who had both sites treated or partially treated with drills, the difference was still significant (P = 0.001).

• Patient preference at day 3: four patients preferred piezoelectric surgery while 16 patients found both techniques equally acceptable; the difference in patient preference between the two surgeries was not statistically significant (P = 0.125). By remov-ing the two patients who had both sites treated or partially treated with drills, the difference was still not significant (P = 0.500).

The comparison of the clinical outcome between the two clinicians is presented in Table 2. There were no apparent differences between the two surgeons.

Discussion

This trial was designed to understand whether it would be more convenient to use piezoelectric surgery or conventional rotational drills to prepare

Table 2 Comparison of the clinical outcomes of the two operators. Each operator treated 10 patients.

Dr Felice Dr Pistilli P = value

Patients with failed prostheses 0 1 1.000§

Patients with failed implants 0 1 1.000§

Patients with complications 5 2* 0.350§

Implant insertion time in min (mean ± SD) 19.45 ± 2.39 18.40 ± 0.91 0.248^

Patients with implants not loaded immediately 0 1 1.000§

Patients with a larger haematoma at drilled sites 5 6 1.000§

Patients preferring piezoelectric surgery 3 1 0.582§

SD: standard deviation; * Patients with bilateral complications; § two-tailed Fisher’s exact test; ^ Mann-Whitney U test

Fig 8a to c Implant protruding from the zygoma (piezo site), the patient reported discomfort and that she could not sleep on that side. a) a flap was elevated under the lower eyelid as a blepharoplasty showing the implant apex extruding outside the zygoma by four threads; b) the protruding portion of the implant was removed with a diamond ball bur; c) clinical view at 11 months after removal of the implant apex, no scar or intervention marks are visible.

cba



sites for zygomatic implants. A split-mouth design was preferred to minimise anatomical variations and other patient-related factors, which could potentially influence the results. Both procedures achieved simi-lar 1-year post-loading results, however some dif-ferences could be noted, despite the limited sample size. The time necessary to place both implants was 9 min shorter using drills and in two patients drills also had to be used on the piezoelectric surgery side since the piezoelectric surgery inserts – used at maximum power – were unable to penetrate the maxillary bone. On the other hand, larger haema-tomas were seen at drilled side, suggesting a more aggressive behaviour of the drills. Complications also tended to be more frequent at drilled sites. Of par-ticular interest were the lips burns (Figs 6a to c) and the fracture of the zygoma (Fig 5) caused by the high insertion torque because of the undersized prepar-ation of the implant sites.

The reason oncology implants were used for this study rather than fully threaded conventional zygo-matic implants was dictated by the author’s prefer-ence to have an unthreaded coronal portion in the area more likely to be exposed to bacterial plaque of the oral cavity over time, to facilitate oral hygiene procedures over the exposed implant surface, in the case of complications such as tissue fenestrations around the implants. One case of implant exposure did occur (Fig 7), however the area remained plaque and inflammation-free for the first year in function, suggesting this was a sound decision.

Expressing a preference, the authors would pre-fer to use rotational drills, however this pilot trial suggests modifications and improvements to both manufacturers. In the opinion of the authors the rotational drills would benefit from an additional drill with a diameter superior to 3.4 mm or a tapping drill for placing 4.3 mm diameter implants, especially in zygoma of very dense bone, to minimise the risk of zygoma fracture (Fig 5). We agree that implant sites should be underprepared to achieve high insertion torques to allow immediate loading, but at the same time it is essential to minimise the risk of zygoma fracture when delivering high torques at implant insertion. We also suggest the production of zygo-matic oncology implants with smaller diameters (if the implant resistance at fracture will not be com-promised) to facilitate the placement of pair implants

in small zygomas. Regarding the piezoelectric sur-gery options, both speed and penetrating capacity into the bone are issues that should be improved to make piezoelectric surgery really competitive with rotational drills. Finally, there is no need to have the implant tip protruding a few mm outside the zygoma. On the contrary this can cause discomfort as happened to one of our patients (Figs 8a to c), so a good stock of zygomatic implants of varying lengths is a necessity to avoid rescue solutions as we had to shorten one of the implants in another patient (Fig 4).

It is interesting to observe that 1 year after load-ing not a single patient was rehabilitated with a de-finitive prosthesis as planned and agreed at the pro-tocol stage. The explanation is that while patients received free implants, they had to pay for their own prostheses. Most likely patients were pleased with the provisional prostheses or had financial problems so they decided to postpone the time to receive the definitive prosthesis. Only two patients were under-going replacement with the definitive prostheses at the time of writing this paper.

It is not possible to compare the present results with those of similar RCTs, since there none are avail-able.

The main limitation of the present investigation was the small number of included patients, however sufficient to provide some useful indications and able to generate hypotheses for future investigations.

Both implant preparation techniques were tested in real clinical conditions and patient inclusion criteria was broad, therefore the results of the present trial can be generalised to a larger population with simi-lar characteristics, keeping in mind that the place-ment of zygomatic implants is a complex procedure requiring skilled and experienced operators, since potentially severe complications may occur.

Conclusions

Both drilling techniques achieved similar clinical results, but conventional drilling required 9 min less and could be used in all instances, although it was more aggressive. These results may be system-dependent, however, so they cannot be generalised to other zygomatic systems with confidence.



References

1. Brånemark PI, Gröndahl K, Öhrnell LO, Nilsson P, Petru-son B, Svensson B, Engstrand P, Nannmark U. Zygoma fix-ture in the management of advanced atrophy of the maxilla: technique and long-term results. Scand J Plast and Reconstr Surg 2004;38: 70–85.

2. Aparicio C. A proposed classification for zygomatic implant patient based on the zygoma anatomy guided approach (ZAGA): a cross-sectional survey. Eur J Oral Implantol 2011; 4: 269–275.

3. Davó R, Pons O. Prostheses supported by four immediately loaded zygomatic implants: a 3-year prospective study. Eur J Oral Implantol 2013;6: 263–269.

4. Esposito M, Felice P, Worthington HV. Interventions for replacing missing teeth: augmentation procedures of the maxillary sinus. Cochrane Database Syst Rev 2014: Chich-ester, UK: John Wiley & Sons, Ltd.

5. Esposito M, Barausse C, Pistilli R, Sammartino G, Grandi G, Felice P. Short implants versus bone augmentation for plac-ing longer implants in atrophic maxillae: One-year post-loading results of a pilot randomised controlled trial. Euro J Oral Implantol 2015;8: 257–268.

6. Stevenson AR, Austin BW. Zygomatic fixtures – the Sydney experience. Ann R Australas Coll of Dent Surg 2000;15: 337–339.

7. Triplett RG, Schow SR, Laskin DM. Oral and maxillofa-cial surgery advances in implant dentistry. The Interna-tional Journal of Oral and Maxillofacial Implants 2000;15: 47–55.

8. Malevez C, Abarca M, Durdu F, Daelemans P. Clinical out-come of 103 consecutive zygomatic implants: a 6–48 months follow-up study. Clin Oral Implants Res 2004;15: 18–22.

9. Hirsch JM, Öhrnell LO, Henry PJ, Andreasson L, Bråne-mark PI, Chiapasco M, Gynther G, Finne K, Higuchi KW, Isaksson S, Kahnberg KE, Malevez C, Neukam FW, Sevetz E, Urgell JP, Widmark G, Bolind P. A clinical evaluation of the Zygoma fixture: one year of follow-up at 16 clinics. J Oral Maxillofac Surg 2004;62: 22–29.

10. Esposito M, Worthington HV. Interventions for replacing missing teeth: dental implants in zygomatic bone for the rehabilitation of the severely deficient edentulous maxilla. Cochrane Database Syst Revs 2013: Chichester, UK: John Wiley & Sons, Ltd.

conventional drills vs piezoelectric surgery preparation

Documents