Quintessence Journals

RANDOMISED CONTROLLED CLINICAL TRIAL 233 pyri Co gh Not for Publicationt

b y Q

i N

n o

t t r f ess o ce Mohamed Badr, Paul Coulthard, Rami Alissa, Richard Oliver en The efﬁcacy of platelet-rich plasma in grafted maxillae. A randomised clinical trial

Mohamed Badr, BDS, MSc Doctoral Researcher, Oral and Maxillofacial Surgery, School of Dentistry, Man- chester Academic Health Science Centre, The University of Manchester, Manchester, UK Key words bone graft, iliac crest, implant stability, maxilla, platelet-rich plasma Paul Coulthard, BDS, MDS, PhD Conflict-of-interest statement: There is no financial conflict of interest. Professor in Oral and Maxil- lofacial Surgery and Editor, Cochrane Oral Health Purpose: The aim of this randomised, controlled, parallel-group clinical trial was to evaluate the clini- Group, School of Dentistry, The University of Manches- cal effect of platelet-rich plasma (PRP) on bone graft healing and implant integration in iliac crest ter, Manchester, UK grafted maxillae. Rami Alissa, PhD Materials and methods: Twenty-two consenting patients were randomised to PRP (13 patients) and Clinical Assistant, Oral and control (9 patients) groups. Both groups received onlays and 16 patients had their maxillary sinus Maxillofacial Surgery, School of Dentistry, Manchester grafted with particulate bone (lateral window approach) with iliac crest bone grafts. Autologous Academic Health Science platelet concentrates were prepared from the patients’ blood and autologous thrombin was pro- Centre, The University of Manchester, Manchester, duced. PRP was mixed with the test group bone grafts. Outcome measures were implant integration, UK implant stability, soft tissue healing, graft resorption, and donor and recipient site complications. Richard J Oliver, Implant stability measurements were recorded at placement and exposure using a resonance fre- BDS, BSc, PhD Private practice, quency analysis device. Patients were followed up to abutment connection. Multiple linear regression Shrewsbury, UK analyses using robust standard error were performed, taking the patient as the unit of measurement. Correspondence to: The t test was also used where appropriate. Mohamed Badr Results: One bone graft failed in the PRP group and regrafting was required. No statistically signifi- School of Dentistry, The University of Manchester, cant differences were observed for soft tissue healing indices (P = 0.4) and mean graft resorption Higher Cambridge Street, (P = 0.5) between groups. All implants were found clinically integrated at time of exposure. No statis- Manchester, UK M15 6FH Email: tically significant differences in implant stability were observed between groups at implant placement [email protected] (P = 0.059) and exposure (P = 0.1). Using a post-hoc analysis, posterior implants in the PRP group showed statistically significantly higher stability values (61 ± 2.6) than anterior implants (60 ± 2.4) at implant placement (mean difference -0.95, P = 0.04). However, this difference was not clinically significant. Conclusion: No appreciable clinical effect could be observed when using PRP with autologous iliac crest bone graft in the maxilla.

Eur J Oral Implantol 2010;3(3):233–244 234 Badr et al Platelet-rich plasma in grafted maxillae pyri Co gh Not for Publicationt

b y Introduction Although the surplus platelet-poorQ plasma (PPP)

available after PRP preparation is expected to containN

n o

Bone volume and quality significantly affect the a significantly lower number of plateletst and conse-t r f ess o e clinical outcome of dental implants together with quently lower levels of growth factors, it shouldenc con- other implant-related factors1. Losing teeth for any tain a fairly high concentration of fibrinogen. Fibrin, reason usually results in losing the supporting bone which is the activated form of fibrinogen, is known as well. Insufficient bone volume often necessitates to play an important role in wound healing. Not only augmenting the site before implant placement and does the fibrin network contribute to haemostasis is frequently undertaken using an autologous bone after capillary injury, but it also acts as an extracellu- graft. However, the need for a second surgical site, lar matrix which is essential for cell migration during with the donor site associated comorbidity, together angiogenesis and other wound-healing processes14. with the inherent tendency for graft resorption dur- Nonetheless, the clinical efficacy of fibrin sealants on ing healing might limit this reconstructive option2. wound healing is still controversial15. One of the most important factors influencing The aim of the present study was to investigate implant success rates is the stability of the implant the effect of PRP on autologous bone graft heal- immediately after insertion, which is referred to as ing and implant stability. The authors carried out a the primary stability3. The primary stability can also randomised clinical trial to test the null hypothesis influence the implant optimal loading time4. An of no differences in hard and soft tissue healing, estimate of the degree of implant stability can be implant stability, and complications, using PRP or performed by measuring insertion torque or using not, against the alternative hypothesis of a differ- Periotest, as well as by a resonance frequency analy- ence between the two procedures. sis (RFA) device which detects the implant’s resonance frequency and translates this into an implant stability quotient (ISQ). The higher the ISQ value, Materials and methods the more stable the implant is expected to be5,6. Lower implant stability and lower success rates were The CONSORT guidelines16 were followed for observed in less dense and/or in grafted bone and reporting the present study. It was a balanced ran- have been reported in a number of studies7,8. domised, controlled, parallel-group trial. The proto- Recently introduced regenerative treatment col of the present study was reviewed and approved options using biological mediators have gained inter- by the Tameside and Glossop Local Research Ethics est from oral and maxillofacial surgeons. Amongst Committee (Ref.No.05/Q1402/29). This study was those biological mediators are the recombinant part of a research degree course. bone morphogenic proteins and the platelet-derived growth factor preparations; both are now commer- Patient recruitment cially available9. However, the technique of human recombinant growth factor preparation is quite Eligible patients were adults with a deficient maxil- expensive and has led to considering more cost- lary ridge who were to receive dental implants for effective alternatives10. fixed or removable prosthetic reconstruction. Eligi- Platelets are considered a rich source of growth ble patients were to receive maxillary augmentation factors which are stored in their alpha granules and using an autologous iliac crest bone graft. Patients released upon platelet activation in areas of tissue were to be excluded from the study if they had any damage11. Platelet-rich plasma (PRP) is an economic of the following: i) abnormal platelet count (less than autologous source for such growth factors and is 150,000 or more than 400,000 platelets/mm3); ii) believed to enhance soft and hard tissue healing, any bleeding and/or clotting disorder (e.g. haemo- leading to a faster graft consolidation11,12. It is also philia, platelet dysfunction syndrome); iii) allergy to thought that it might lead to a lower degree of bone tetracycline (in order to be suitable for bone label- resorption, which is considered a main shortcoming ling procedure); iv) regular or recent aspirin intake frequently seen in autologous bone grafts13. (within the last 10 days); v) heavy smokers (more

Eur J Oral Implantol 2010;3(3):233–244 Badr et al Platelet-rich plasma in grafted maxillae 235 pyri Co gh Not for Publicationt

Fig 1 PRP (bottom,y than 20 cigarettes/day); vi) any general medical con- Q

with noozle) and PPP

u dition that contraindicates general anaesthesia; vii) N i (top, in tray) loaded in

n o

the applicators provided under treatment with NSAIDs, antibiotics, systemic t t r f e o bys these manufacturer.nce corticosteroids, anticoagulants or immunosuppres- The applicator provides sive drugs; viii) neoplasia or haematological malig- the PRP/PPP and nancy; ix) metabolic bone disease such as osteoma- thrombin in a 10:1 ratio, respectively. lacia, hypocalcaemia or hypercalcaemia, x) Jehovah’s Witness; or xi) participation in another trial. Patients were referred to the Oral and Maxillo- facial Surgery Unit, University Dental Hospital of Manchester. From the period of July 2007 to Janu- ary 2010, patients were approached and asked to participate in the present study. Patients were asked to sign an informed consent form and, using a com- width augmentation. Grafts were fixed using tita- puterised random allocation process, patients were nium screws (diameter 1.5 mm, Martin Medizintech- allocated to either a test or control group. nik, Tuttlingen, Germany) to ensure graft immobility In the test group patients, PRP was mixed with the during the healing period. bone graft during and after graft fixation. Patients’ allocation was placed in a sealed opaque envelope PRP preparation by the department secretary. Immediately before the induction of anaesthesia for the bone graft surgery, For patients in the test group, PRP was prepared at the envelope corresponding to the current participant the same time as graft harvesting by one of the study was opened and the patient’s group was revealed investigators (MB). by Mohamed Badr (MB). All patients were blind to After anaesthetic induction, 54 ml of blood was their group of study. All surgeries were undertaken drawn using a green aphaeresis needle into a 60 ml by one of two surgeons (Richard Oliver [RO] or Paul syringe pre-filled with 6 ml of anticoagulant citrate Coulthard [PC]) and their surgical teams. dextrose solution. The syringe was gently agitated The need for onlay with or without inlay maxil- several times to insure even distribution of the anti- lary bone grafting was mandatory for patient par- coagulant. The citrated blood was then loaded into ticipation. All patients received a preoperative pano- GPS®II canisters (Gravitational Platelet Separation ramic radiograph and study models for treatment II, Biomet Biologics, Warsaw, IN, USA). Filled can- planning, stent fabrication and for sinus anatomy isters were placed in a specially designed centrifuge assessment. (Biomet Biologics centrifuge, 120 V, 50 to 60 Hz) with a balancing load and spun for 15 minutes at 3200 rpm according to the manufacturer’s instruc- Graft harvesting tions (Biomet GPSII User Manual, 2008). This proc- All patients were admitted as in-patients in the Man- ess separated the blood into red blood cells, PPP, chester Royal Infirmary. Under general anaesthesia, and buffy coat components, the latter contained the cortico-cancellous block grafts were harvested from PRP and white blood cells. Autologous thrombin was the medial aspect of the anterior iliac crest for all then prepared using a thrombin processing device patients. The harvested graft was adjusted with burs (TPD™, ThermoGenesis, Rancho Cordova, CA, under copious irrigation to adapt to the recipient USA) utilising the patient’s own PPP in a process that surface. For sinus inlay augmentations, cancellous took about 25 minutes. PRP and thrombin were then as well as milled cortico-cancellous bone grafts were loaded in a special applicator (FibriJet®, Micromed- carefully packed under the elevated sinus lining ics, Eagan, MN, USA) which delivered them in a 10:1 which was lifted following a standard lateral win- ratio, respectively, for PRP activation. The remaining dow approach17,18. Onlay bone grafts were used PPP and thrombin were loaded in a similar applicator to replace the lateral wall of the sinus and for ridge for soft tissue application (Fig 1).

Eur J Oral Implantol 2010;3(3):233–244 236 Badr et al Platelet-rich plasma in grafted maxillae pyri Co gh Not for Publicationt

y PRP application Outcome measures Q

i N

n o

PRP was applied to the cancellous bone and packed The following outcome measures weret examined.t r f ess o e into the 3D space created from the sinus lining eleva- enc tion. For onlay block grafts, PRP was sprayed on the Primary outcome measures prepared recipient bed, through the drilled screw holes and all over the fixed block before flap sutur- Implant integration ing. PPP was then sprayed on sutured flaps for accel- Successful implant integration was investigated as eration of soft tissue healing. At all times, care was the primary outcome measure. Implant integration taken to avoid washing out or aspirating the formed was clinically assessed by applying manual torque to PRP/PPP clot after their application. All patients the implants’ cover screw at abutment connection were provided with chlorhexidine mouthwash after surgery. implant placement and were instructed to use the mouthwash for 2 weeks. Secondary outcome measures Ridge mapping Implant placement and exposure The ridge width was measured before and after Implant placement was performed 3 to 4 months after onlay graft fixation for both groups. A Boley gauge bone augmentation. After anaesthetic administration, calliper (General Medical, Wiltshire, UK), with the all fixation screws were removed and a crestal inci- guidance of a prefabricated acrylic stent, was used sion was performed to minimally expose the bony at reproducible points. This measurement was ridge. With the help of the fabricated acrylic stent, repeated 1 month later and at implant placement the implant placement positions were marked using for both groups. Each measurement was taken a small round bur. Surgical trephine burs (Frios®, twice, at the beginning and at the end of the fol- Friadent, Mannheim, Germany) with internal diam- low-up appointments, and an average measure- eter of 2 or 2.5 mm were used to harvest bone core ment was recorded. samples from the implant placement sites for histomorphometric analysis. After harvesting the bone Soft tissue assessment core, the osteotomy site was prepared as usual follow- Two weeks after surgery, a soft tissue healing index ing the implant system manufacturer recommenda- was recorded for both control and test groups. The tions (OsseoSpeed™, Astra Tech, Mölndal, Sweden). following criteria were used: score 4 = excellent heal- All implants were placed at the level of the bone ridge. ing with pink gingiva, no bleeding, no granulation Resorbable intra-oral sutures (3-0 Vicryl, Ethicon) tissue, and no dehiscence; score 3 = good healing with were used in all patients; any remaining sutures were only slight to moderate gingival redness; score 2 = removed by the second week. Removable prostheses poor wound healing with significant erythema, bleed- were fitted and as necessary relined using chairside ing on palpation, granulation or dehiscence; score 1 = relining material (Visco-gel, Dentsply DeTrey, Kon- very poor healing with suppuration and/or spontane- stanz, Germany) and patients were instructed not to ous bleeding. This healing index is a modification of start using their prostheses before the end of the sec- a previously published wound healing index (WHI) ond week following bone graft surgery. used to evaluate the coronally advanced flap root Five to 6 months were allowed for the maxillary coverage procedure19. The soft tissue assessment was implants to integrate. At abutment connection, a undertaken by an independent assessor blinded to the small crestal incision was used to expose the implants, patients’ treatment. cover screws were removed and healing abutments (Astra Tech) with a suitable height were placed after Bone labelling the implant stability readings were taken. Resorba- All patients were given tetracycline antibiotic for ble intra-oral sutures (3-0 Vicryl, Ethicon) were used bone labelling. Each patient was required to take when necessary. two doses of tetracycline with a gap of 10 days

Eur J Oral Implantol 2010;3(3):233–244 Badr et al Platelet-rich plasma in grafted maxillae 237 pyri Co gh Not for Publicationt

Fig 2 Osstell Mentory in between. The last dose was 4 days prior to the Q

device used for implant

u implant placement appointment and each dose con- N i stability measurement

n o

showing an ISQ reading sisted of 2 g of tetracycline. Each dose was split into t t r f e o of s73.s e two sub-doses of 1 g each, taken 12 hours apart. At enc implant placement, a bone core was harvested from the implant placement site for static and dynamic histomorphometric analyses. The completed results for the histomorphometric analysis will be presented in a future report.

Implant stability measurement The ISQ was taken immediately after implant placement using an RFA device (Osstell Mentor, Osstell, Göteborg, Sweden). This updated version of the device consists of a SmartPeg which is manually screwed into the implant using a plastic mount. The USA) for comparing the difference in mean bone measurement probe is used to excite the SmartPeg resorption between groups. The Mann–Whitney U by a magnetic pulse, then it records the response and test was used for analysing the soft tissue heal- translates this in the form of an ISQ on a scale from 1 ing index results. Statistical significance was set at to 100, the latter being the highest measurable stabil- 5%. An independent t test was used to investigate ity quotient20. The process is wireless with no need for the operators’ effect on bone resorption, soft tissue any physical contact between the device parts. This healing and implant stability. measurement was repeated at exposure and healing An implant stability comparison between the PRP abutment connection stage. The readings were taken and control groups for anterior and posterior maxil- from a bucco-lingual as well as a mesio-distal direc- lae was performed as a post-hoc analysis. tion, and the average ISQ was calculated (Fig 2). Only the soft tissue healing index was recorded by a blinded assessor. Statistics The sample size calculation in the present study was Results based on implant stability (secondary outcome). A power calculation based sample size of 15 patients A flow diagram of the progress through the phases per group was the target of the study. This calcu- of the trial can be seen in Figure 3. lation had 90% power to detect a 10% increase The rate of patient recruitment in the unit in the implant stability measurement. The Stata resulted in a recruitment of 22 patients in a period of (Intercooled Stata 9.2 for Windows, StatCorp, Col- 3 years. One female patient who was not medically lege Station, TX, USA) statistical package was used fit for general anaesthesia was ineligible for the study. for calculating the sample size. This was based on As the present study was part of a degree course, maxillary implant stability measurements reported patient recruitment had to stop to allow for the com- in a previously published study21 in which the mean pletion of the patients’ follow-up by the end of the implant stability in the control group was 55 and the course. Patients recruited in the study (14 females, common standard deviation was 3.8. 8 males) had a mean age of 36 years (range 17 to Clustering of implants within patients was 73). All patients had onlay block grafts to the anterior considered in all implant stability measurements. maxilla, while 16 patients, in addition to onlay block Multiple linear regression analyses using robust grafts, also needed maxillary sinus augmentation standard error were used for stability analyses (3 unilateral, 13 bilateral) using inlay bone grafting (Stata). The t test was performed using the SPSS at the same time. A total of 85 implants were placed software package (version 16, SPSS, Chicago, IL, in 22 patients. A total of 48 implants were placed

Eur J Oral Implantol 2010;3(3):233–244 238 Badr et al Platelet-rich plasma in grafted maxillae pyri Co gh Not for Publicationt

Fig 3 A ﬂow diagram y Q of the progress through

u Assessed for eligibility N the different trial i

n o

phases. (n = 23) t t r f e o ssence Enrolment Allocation Follow-up

Excluded (n = 1) Not meeting inclusion criteria (n = 1) The patient was not ﬁt for surgery under general anaesthesia

Randomised (n = 22)

Allocated to PRP (n = 13) Allocated to control group (n = 9) Received allocated PRP (n = 13) Received standard treatment (n = 9)

Lost to follow-up (n = 0) Lost to follow-up (n = 0) Discontinued intervention (n = 0) Discontinued intervention (n = 0) Analysis

Analysed (n = 12) Analysed (n = 9) Data not available (n = 1) Excluded from analysis (n = 0) This patient needed further bone grafting and implant placement was deferred

Table 1 Baseline characteristics of the study patients. in the maxilla of 13 patients who had received their bone grafts mixed with PRP, and 37 implants were Variable PRP Control n = 13 n = 9 placed in the maxilla of nine control-group patients. The implant diameters ranged from 3.5 to 5 mm, and Age (years) 38 32 lengths ranged from 9 to 19 mm (Table 1). Females 7 7

Surgeon (PC) 8 3 Platelet yield

Surgeon (RO) 5 6 The PRP preparation system achieved a 4- to 7-fold (5.4 ± 0.9) increase in platelet concentration above Maxillary anterior 27 13 the baseline level with up to 78% platelet recovery implants rate. Maxillary posterior 21 24 implants

Mean implant length Clinical outcomes 14 14 (mm) The donor site healed in all cases with no signs of Mean implant diam- 44 infection. One case developed a mild haematoma eter (mm) which resolved gradually. No sensory disturbances

Eur J Oral Implantol 2010;3(3):233–244 Badr et al Platelet-rich plasma in grafted maxillae 239 pyri Co gh Not for Publicationt

Table 2 Gingival healing index scores (1 = very poor healing, 4 = excellent healing). y Q

Gingival healing index 1234Total i N

n o

t t r f e o PRP 036413 ssence Control 01449

Total 0 4 10 8 22

Table 3 Distribution of mean graft resorption in PRP and control groups.

Resorption (mm) Frequency in control Frequency in PRP Total

0.5 0 1 1

0.8 1 1 2

1011

1.1 1 1 2

1.3 0 2 2

1.4 1 2 3

1.5 2 0 2

1.6 0 1 1

1.8 1 0 1

1.9 0 2 2

2213

2.5 1 0 1

2.6 0 1 1

Total 9 13 22

were recorded in any of the 22 patients. All patients Soft tissue healing and graft resorption were able to attend the first follow-up and walked into the surgery unassisted. All skin sutures were Gingival healing indices showed no significant dif- removed after 7 to 10 days. ference between test and control groups at 2 weeks Only two postoperative complications occured, after bone graft surgery (P = 0.4). The mean gingival one each group. One patient (PRP group) who had healing index at 2 weeks after bone graft surgery an extensively resorbed maxilla, had an antral com- was 3.1 ± 0.75 and 3.3 ± 0.7 for the PRP and control munication about 4 weeks after her bone graft sur- groups, respectively (Table 2). The independent t test gery which was corrected under general anaesthe- did not reveal any significant differences in the mean sia. This patient needed further bone grafting and gingival healing index scores between the two sur- therefore her implant placement was deferred and geons (3.0 ± 0.7 and 3.4 ± 0.6 mean wound healing subsequent implant stability measurements were index for operators 1 and 2, respectively; P = 0.2). not available for analysis. A small area of soft tissue The ridge mapping procedure revealed minor wound dehiscence occurred in one patient (control graft resorption in both PRP and control groups. An group) that did not need intervention and healed overall range of graft resorption from 0.5 to 2.5 mm spontaneously with secondary intention. was evident in the present study. The patient that required regrafting showed 1.3 mm of resorption. No significant difference between PRP and con-

Eur J Oral Implantol 2010;3(3):233–244 240 Badr et al Platelet-rich plasma in grafted maxillae pyri Co gh Not for Publicationt

Table 4 Mean implant stability quotients at implant placement. y Q

PRP Group Control Group P value Mean 95% Conﬁdence intervali of the N

n o

(n = 12) (n = 9) difference difference t t r f e o ssence Lower Upper

Maxilla 63 ± 2.2 61.8 ± 4.2 0.059 1.95 -0.08 3.99

Maxillary anterior 63 ± 3.3 64 ± 4.8 0.989 -0.013 -2.07 2.01

Maxillary posterior 61 ± 2.6 60 ± 2.4 0.04* -0.95 -1.84 -0.049

*Statistically signiﬁcant difference

Table 5 Mean implant stability quotients at abutment connection.

PRP Group Control P value Mean 95% Conﬁdence interval of the (n = 12) Group (n = 9) difference difference

Lower Upper

Maxilla 65 ± 3.3 64 ± 4.5 0.129 2.46 -0.78 5.69

Maxillary anterior 66 ± 3.6 64 ± 7.1 0.097 -2.27 -4.99 0.46

Maxillary posterior 65 ± 2.2 64 ± 1.7 0.627 -0.34 -1.84 1.15

trol groups was found in the average amount of differences in the mean implant stability measure- graft resorption measured by the surgical callipers ments were observed between groups at abutment (1.4 ± 0.5 and 1.6 ± 0.5, respectively; P = 0.5). The connection stage (P = 0.1, Table 5) mean graft resorption and its distribution in both Implant stability quotients for the posterior max- PRP and control patients are presented in Table 3. No illary implants in the PRP group showed statistically significant difference was observed in the mean graft significantly higher mean values compared to the same resorption between the two surgeons (1.4 ± 0.6 and quadrant in the control group at implant placement 1.5 ± 0.3 mean resorption in mm for operators 1 (61.6 ± 2.6 and 60 ± 2.4, respectively, P = 0.04) (Fig 4). and 2, respectively; P = 0.5) The mean ISQ at abutment connection was statis- Apart from the bone regrafting case, implant tically significantly higher than at implant placement placement was undertaken as planned. However, a (64.6 ± 4.3 and 61.8 ± 4.3, respectively, P < 0.002). modification in implant angulation and/or implant Comparable implant stability values were obtained diameter was occasionally necessary to avoid bone by both operators at implant placement stage dehiscence during implant site preparation. (63 ± 3.6 and 62 ± 2.4, operators 1 and 2, respectively; P = 0.6, 95% confidence interval -2.04 to 3.6). Implant integration All implants were found to be clinically integrated Discussion at time of exposure and abutment connection, with no signs of mobility or infection in either of the two The present study failed to show any statistically groups. or clinically significant differences when using PRP with autologous grafts in maxillae. This may be due to the fact that PRP added no clinical ben- Implant stability measurements efit in this application, or maybe that the sample No significant differences in the mean implant stabil- size was insufficient to detect a possible difference. ity measurements were observed between groups at Ideally, sample size calculations should be based implant insertion (P = 0.059, Table 4). No significant on a primary outcome such as successful implant

Eur J Oral Implantol 2010;3(3):233–244 Badr et al Platelet-rich plasma in grafted maxillae 241 pyri Co gh Not for Publicationt

b y integration. This was expected to result in a sample Q

size of hundreds if not thousands of patients, which N

n o

was not feasible for the present study’s recruitment t t r f ess o e capacities. However, the present data can still be enc useful when gathered with data from other studies in systematic reviews. A number of studies showing either positive or no effect of PRP application on bone and/or soft tissue healing have been published22-26. A limited number of studies were randomised controlled trials. Moreover, only a few of them discussed the clinical significance of the possible biological effect of PRP on bone and soft tissue healing27-29. The authors aimed in the present study to investigate the clinical significance of the possible favourable biological effect of PRP in a well-designed clinical trial as recommended by previous reviewers30,31. Assessing soft tissue healing, bone resorption, Fig 4 Box plot showing the average ISQ values at implant placement. UA = maxillary implant stability measurements as well as implant anterior, UP = maxillary posterior. A significant difference was noted between posterior placement ideally should be performed by a blinded maxilla quadrants in the PRP and control groups, P = 0.04. assessor and/or operator. However, due to the available resources during the study, only the soft tissue values, which possibly leaves little influence for the healing index was recorded by a blinded assessor. healing process on future implant stability. Therefore, In the present clinical experiment, all implants the authors recommended further investigations on were found to be clinically integrated at abutment the effect of PRP on less dense bone areas. connection with no preloading failures in either of Although the present authors observed no signif- the two groups. Comparable wound healing pat- icant differences in implant stability values between terns and graft resorption was observed between groups, a statistically significantly higher mean ISQ groups. Although slightly higher ISQ values were was observed in the maxillary posterior implants observed in the posterior maxilla quadrants in the in the test group patients compared to the control PRP group, no significant differences were observed group at implant placement stage. This statistical in the mean ISQ between groups at both placement significance should be interpreted with care, espe- and abutment connection stages. cially with the small number of patients recruited in The RFA device is a non-invasive method for the study and the trivial, if any, clinical benefit this quantifying the degree of implant stability. Previous statistical significance might indicate. publications have noted that RFA devices might not The significant difference was no longer found at be reliable for detecting implant instability; however, abutment connection, and similar ISQ values were they have also demonstrated that RFA devices could observed between maxillary quadrants as well as reliably indicate if the implant under investigation is between maxillae in the PRP and control groups. This clinically stable32. At the same time, a positive cor- was in agreement with Friberg et al, who reported relation between bone density and the degree of improved ISQ values over time for implants placed in implant stability was observed by Turkyilmaz and less dense bone areas35. The same was also reported McGlumphy33. by Sjostrom et al in their study investigating implant Few studies have tried to investigate the effect of stability placed in grafted bone, where they observed PRP on implant stability. Monov et al found no addi- a levelling take place over time between two extreme tional benefits of PRP on implant stability in the man- ISQ values36. This might further support the possibil- dible34. They attributed the insignificant effect of PRP ity that PRP might not have a clinically appreciable to the high density of mandibular bones with high ISQ effect on the implant integration process.

Eur J Oral Implantol 2010;3(3):233–244 242 Badr et al Platelet-rich plasma in grafted maxillae pyri Co gh Not for Publicationt

b y In a split-mouth study, Thor et al reported a sig- authors assumed that no other clinicalQ variable was

niﬁcant difference in ISQ values between PRP and measured in the study. These results can be con-N

n o

non-PRP sides. By further analysing their results, the sidered to be in agreement with the tpresent resultst r f ess o e authors reported that they found the statistically sig- as Lindeboom et al reported no clinically significantenc nificantly higher values in the PRP group only in the difference after 2 weeks of healing. This could be a anterior area of the maxilla. Since two different graft result of the fast healing of intra-oral wounds, pos- types were used in each side of the anterior area, the sibly due to high intra-oral vascularisation41. authors could not conclusively attribute the differ- A positive effect of PPP (fibrin glue) on healing ence to the effect of the PRP, but rather suggested intraoral mucosal wounds could not be confirmed in that the effect was related to the graft type27. the present study. Since PRP has been shown to be The results from the present study seem to be in a significant stimulus for tissue repair in chronic non- agreement with previous results of no effect, spe- healing cutaneous ulcers37 as well as in the manage- cifically those reported in a recent systematic review ment of acute trauma wounds42, PRP application which concluded that PRP treatment does not seem to therefore might be of greatest benefit in problematic improve the clinical outcome of sinus lift procedures26. areas where healing enhancement is desirable, such as A number of clinical as well as experimental stud- in immunocompromised patients and in chronic ulcers. ies have reported promising results for the effect of Bone graft resorption is one of the most chal- fibrin sealants and PRP preparations on wound heal- lenging difficulties facing autologous bone augmen- ing and haemostasis in intra-oral as well as extra-oral tation procedures. In the present study an average wounds14,37-39. Clinically, Yoo et al reported a better of 1.5 ± 0.5 mm of bone resorption was detected wound healing effect of PRP and PPP application from the graft width, which was in agreement with following hemithyroidectomy in a randomised con- the amount of bone resorption observed by Chia- trolled trial by showing a 29.3% drainage reduc- pasco et al43. However, a more significant resorption tion in the PRP group38. Histologically, Yucel et al has also been reported, which occasionally led to observed a positive effect of fibrin glue on the heal- changes in the planned implant placement site and ing of extraction sockets in a rat model14. less frequently necessitates bone regrafting44. In the A tangible favourable clinical effect for PPP how- present study, one case from the PRP group required ever, could not be confirmed in the present study. A regrafting before implant placement. The majority possible explanation is that haemostasis, which was a of the resorbed bone in this patient was the par- key factor in assessing the favourable effect of fibrin ticulated bone used to augment the sinuses, while glue in the previous studies was not measured in the the onlay bone graft showed comparable amount of present study as bleeding is not normally a concern bone resorption to the rest of the study cases. during this procedure. At the same time, reduction in Both the PRP and control patients showed simi- drainage, which is another important parameter utilised lar amounts of bone resorption with no significant as a sign of wound healing enhancement, can be due differences between groups. Taking into considera- to the ability of fibrin glue to seal severed capillaries and tion the method of assessment used in the present not necessarily a sign of accelerated or better healing. study, which measured changes in the ridge width, Orthogonal polarisation spectral (OPS) imag- the PRP did not seem to alter the graft resorption ing is a technique used to assess capillary density rate. The present results are in agreement with Lee et noninvasively by visualising and capturing images al, who observed a comparable graft resorption rate of the gingival microvasculature40. Lindeboom et in autologous grafts with or without PRP in alveolar al observed a significant positive effect of PRP on cleft operations45. On the other hand, PRP in combi- mucosal wound healing after bilateral sinus augmen- nation with surgical-grade calcium sulphate (SGCS) tation surgery as shown by significantly higher capil- was reported by Shi et al to reduce alveolar ridge lary density in the PRP-treated side12. The significant resorption in a canine socket extraction model46. difference in the increase of the capillary density was However, it was also observed in their study that a only notable during the first 10 days after surgery, SGCS/PRP combination showed similar results com- although it was not clearly mentioned. The present pared to SGCS-treated sockets. This might suggest

Eur J Oral Implantol 2010;3(3):233–244 Badr et al Platelet-rich plasma in grafted maxillae 243 pyri Co gh Not for Publicationt

Fig 5 The patterny that SGCS may have played the main role in the Q

of bone graft resorp-

u observed ridge preservation compared to the control N i tion might indicate the

n o

need for a more precise empty sockets, which showed a significantly higher t t r f e o methodsse ntoc assesse graft degree of bone resorption. dimensional changes. The prefabricated acrylic stent may allow for graft width measurement at reproducible points. However, the pattern of graft resorption that was observed in certain cases might suggest that surgical callipers can possibly over- or underestimate the actual amount of graft resorption (Fig 5). Cone- beam computerised tomography scans can be used to detect changes in graft volume, however such a technique entails exposure to a high dose of radia- tion with the potential risks associated with it47. Autologous bone graft resorption still presents a We would like to acknowledge Professor Helen challenge. Although PRP did not show lower resorp- Worthington from the University of Manchester for tion rates, the observed irregular pattern of bone her help in the statistical analyses. resorption might make it more difficult to assess these dimensional changes. An optimised method of measuring graft dimensional changes is yet to be References described. However, the significance of monitoring 1. Esposito M, Hirsch JM, Lekholm U, Thomsen P. Biologi- the graft dimensional changes is debatable if the cal factors contributing to failures of osseointegrated oral implants. (I). Success criteria and epidemiology. Eur J Oral planned optimal implant position can be achieved Sci 1998;106:527-551. despite the anticipated graft resorption. 2. McAllister BS, Haghighat K. Bone augmentation techniques. J Periodontol 2007;78:377-396. While the present study did not show a nota- 3. Lioubavina-Hack N, Lang NP, Karring T. Significance of pri- ble clinical effect of PRP on implant stability dur- mary stability for osseointegration of dental implants. Clin Oral Implants Res 2006;17:244-250. ing the implant integration process, generalisation 4. Cannizzaro G, Leone M, Consolo U, Ferri V, Esposito M. of the present findings should be made with cau- Immediate functional loading of implants placed with flap- less surgery versus conventional implants in partially eden- tion because of the small sample size and the short tulous patients: a 3-year randomized controlled clinical trial. follow-up period. The lack of blinding for graft Int J Oral Maxillofac Implants 2008;23:867-875. 5. Balshi SF, Wolfinger GJ, Balshi TJ. An examination of imme- resorption assessment and implant placement are diately loaded dental implant stability in the diabetic patient also limiting factors in the present study. using resonance frequency analysis (RFA). Quintessence Int 2007;38:271-279. 6. Bischof M, Nedir R, Szmukler-Moncler S, Bernard JP, Samson J. Implant stability measurement of delayed and immediately loaded implants during healing. Clin Oral Implants Res Conclusions 2004;15:529-539. 7. Johansson B, Back T, Hirsch JM. Cutting torque measure- Within the limitations of the present study, the ments in conjunction with implant placement in grafted and nongrafted maxillas as an objective evaluation of bone den- authors can conclude that PRP did not demonstrate sity: a possible method for identifying early implant failures? any clinically significant benefit in maxillae grafted Clin Implant Dent Relat Res 2004;6:9-15. 8. Alsaadi G, Quirynen M, Komarek A, Van Steenberghe D. with autologous iliac bone. Impact of local and systemic factors on the incidence of late oral implant loss. Clin Oral Implants Res 2008;19:670-676. 9. Kao RT, Murakami S, Beirne OR. The use of biologic mediators and tissue engineering in dentistry. Periodontol 2000 Acknowledgements 2009;50:127-153. 10. Kwan MD, Slater BJ, Wan DC, Longaker MT. Cell-based therapies for skeletal regenerative medicine. Hum Mol Genet This research was supported by the Manchester 2008;17:93-98. Academic Health Sciences Centre (MAHSC) and the 11. Marx RE, Carlson ER, Eichstaedt RM, Schimmele SR, Strauss JE, Georgeff KR. Platelet-rich plasma: Growth fac- National Institute for Health Research (NIHR), Man- tor enhancement for bone grafts. Oral Surg Oral Med Oral chester Biomedical Research Centre (BRC). Pathol Oral Radiol Endod 1998;85:638-646.

Eur J Oral Implantol 2010;3(3):233–244 244 Badr et al Platelet-rich plasma in grafted maxillae pyri Co gh Not for Publicationt

12. Lindeboom JA, Mathura KR, Aartman IH, Kroon FH, Milstein 31. Nikolidakis D, Jansen JA. The biology of platelet-rich plasmay Q DM, Ince C. Inﬂuence of the application of platelet-enriched and its application in oral surgery: literature review. Tissue

plasma in oral mucosal wound healing. Clin Oral Implants Eng Part B Rev 2008;14:249-258. i N

n o

Res 2007;18:133-139. 32. Nedir R, Bischof M, Szmukler-Moncler S, Bernard JP, Samson t t r f e o 13. Chiapasco M, Abati S, Romeo E, Vogel G. Clinical outcome J. Predicting osseointegration by means of implantssen primaryce of autogenous bone blocks or guided bone regeneration stability. Clin Oral Implants Res 2004;15:520-528. with e-PTFE membranes for the reconstruction of narrow 33. Turkyilmaz I, McGlumphy EA. Influence of bone density on edentulous ridges. Clin Oral Implants Res 1999;10:278-288. implant stability parameters and implant success: a retro- 14. Yucel EA, Oral O, Olgac V, Oral CK. Effects of fibrin glue on spective clinical study. BMC Oral Health 2008;8:32-40. wound healing in oral cavity. J Dent 2003;31:569-575. 34. Monov G, Fuerst G, Tepper G, Watzak G, Zechner W, 15. Clark RA. Fibrin sealant in wound repair: a systematic sur- Watzek G. The effect of platelet-rich plasma upon implant vey of the literature. Expert Opin Investig Drugs 2000;9: stability measured by resonance frequency analysis in 2371-2392. the lower anterior mandibles. Clin Oral Implants Res 16. Schulz KF, Altman DG, Moher D. CONSORT 2010 State- 2005;16:461-465. ment: updated guidelines for reporting parallel group ran- 35. Friberg B, Sennerby L, Meredith N, Lekholm U. A compari- domised trials. BMC Med 2010;8:18-26. son between cutting torque and resonance frequency meas- 17. Boyne PJ, James RA. Grafting of the maxillary sinus floor with urements of maxillary implants. A 20-month clinical study. autogenous marrow and bone. J Oral Surg 1980;38:613-616. Int J Oral Maxillofac Surg 1999;28:297-303. 18. Tatum H, Jr. Maxillary and sinus implant reconstructions. 36. Sjostrom M, Lundgren S, Nilson H, Sennerby L. Monitoring Dent Clin North Am 1986;30:207-229. of implant stability in grafted bone using resonance frequency 19. Huang LH, Neiva RE, Wang HL. Factors affecting the out- analysis. A clinical study from implant placement to 6 months comes of coronally advanced flap root coverage procedure. of loading. Int J Oral Maxillofac Surg 2005;34:45-51. J Periodontol 2005;76:1729-1734. 37. Mazzucco L, Medici D, Serra M, Panizza R, Rivara G, Orec- 20. Meredith N, Alleyne D, Cawley P. Quantitative determination chia S, Libener R, Cattana E, Levis A, Betta PG, Borzini P. of the stability of the implant-tissue interface using resonance The use of autologous platelet gel to treat difficult-to-heal frequency analysis. Clin Oral Implants Res 1996;7:261-267. wounds: a pilot study. Transfusion 2004;44:1013-1018. 21. Oates TW, Valderrama P, Bischof M, Nedir R, Jones A, Simp- 38. Yoo J, Roth K, Hughes B, Fung K, Franklin J, Lampe H, son J, Toutenburg H, Cochran DL. Enhanced implant stability Pietrzak WS. Evaluation of postoperative drainage with with a chemically modified SLA surface: a randomized pilot application of platelet-rich and platelet-poor plasma fol- study. Int J Oral Maxillofac Implants 2007;22:755-760. lowing hemithyroidectomy: a randomized controlled clinical 22. Philippart P, Brasseur M, Hoyaux D, Pochet R. Human trial. Head Neck 2008;30:1552-1558. recombinant tissue factor, platelet-rich plasma, and tetra- 39. Man D, Plosker H, Winland-Brown JE. The use of autolo- cycilne induce a high-quality human bone graft: a 5-year gous platelet-rich plasma (platelet gel) and autologous survey. Int J Oral Maxillofac Implants 2003;18:411-416. platelet-poor plasma (fibrin glue) in cosmetic surgery. Plast 23. Raghoebar GM, Schortinghuis J, Liem RS, Ruben JL, van Reconstr Surg 2001;107:229-237. der Wal JE, Vissink A. Does platelet-rich plasma promote 40. Lindeboom JA, Mathura KR, Ramsoekh D, Harkisoen S, remodeling of autologous bone grafts used for augmenta- Aartman IH, van den Akker HP, Ince C. The assessment of tion of the maxillary sinus floor? Clin Oral Implants Res the gingival capillary density with orthogonal spectral polari- 2005;16:349-356. zation (OPS) imaging. Arch Oral Biol 2006;51:697-702. 24. Oyama T, Nishimoto S, Tsugawa T, Shimizu F. Efficacy of 41. Szpaderska AM, Zuckerman JD, DiPietro LA. Differential platelet-rich plasma in alveolar bone grafting. J Oral Maxil- injury responses in oral mucosal and cutaneous wounds. J lofac Surg 2004;62:555-558. Dent Res 2003;82:621-626. 25. Kassolis JD, Reynolds MA. Evaluation of the adjunctive ben- 42. Kazakos K, Lyras DN, Verettas D, Tilkeridis K, Tryfonidis M. efits of platelet-rich plasma in subantral sinus augmentation. The use of autologous PRP gel as an aid in the management J Craniofac Surg 2005;16:280-287. of acute trauma wounds. Injury 2009;40:801-805. 26. Esposito M, Grusovin MG, Rees J, Karasoulos D, Felice P, Alis- 43. Chiapasco M, Zaniboni M, Boisco M. Augmentation pro- sa R, Worthington H, Coulthard P. Interventions for replac- cedures for the rehabilitation of deficient edentulous ridges ing missing teeth: augmentation procedures of the maxillary with oral implants. Clin Oral Implants Res 2006;17:136-159. sinus. Cochrane Database Syst Rev 2010;3:D008397. 44. Widmark G, Andersson B, Andrup B, Carlsson GE, Ivanoff 27. Thor A, Wannfors K, Sennerby L, Rasmusson L. Reconstruc- CJ, Lindvall AM. Rehabilitation of patients with severely tion of the severely resorbed maxilla with autogenous bone, resorbed maxillae by means of implants with or without platelet-rich plasma, and implants: 1-year results of a con- bone grafts. A 1-year follow-up study. Int J Oral Maxillofac trolled prospective 5-year study. Clin Implant Dent Relat Res Implants 1998;13:474-482. 2005;7:209-220. 45. Lee C, Nishihara K, Okawachi T, Iwashita Y, Majima HJ, 28. Shepherd N, Greenwell H, Hill M, Vidal R, Scheetz JP. Root Nakamura N. A quantitative radiological assessment of coverage using acellular dermal matrix and comparing a outcomes of autogenous bone graft combined with platelet- coronally positioned tunnel with and without platelet-rich rich plasma in the alveolar cleft. Int J Oral Maxillofac Surg plasma: a pilot study in humans. J Periodontol 2009;80: 2009;38:117-125. 397-404. 46. Shi B, Zhou Y, Wang YN, Cheng XR. Alveolar ridge preserva- 29. Alissa R, Esposito M, Horner K, Oliver R. The Influence of tion prior to implant placement with surgical-grade calcium platelet-rich plasma on the healing of extraction sockets: sulfate and platelet-rich plasma: a pilot study in a canine An explorative randomised clinical trial. Eur J Oral Implantol model. Int J Oral Maxillofac Implants 2008;22:656-665. 2010;3:121-134. 47. Smolka W, Eggensperger N, Carollo V, Ozdoba C, Iizuka T. 30. Arora NS, Ramanayake T, Ren YF, Romanos GE. Platelet- Changes in the volume and density of calvarial split bone Rich Plasma in Sinus Augmentation Procedures: A Systemat- grafts after alveolar ridge augmentation. Clin Oral Implants ic Literature Review: Part II. Implant Dent 2010;19:145-157. Res 2006;17:149-155.

Eur J Oral Implantol 2010;3(3):233–244