Treatment outcomes of implant-retained lower dentures on two endosseous implants placed in severely atrophied mandibles after reconstruction with iliac crest onlay grafts were assessed in a retrospective observational study. All consecutive patients treated between 2000 and 2007 were recalled in 2012 ( n = 40). Survival of the implants, the condition of hard and soft peri-implant tissues, and patient satisfaction were scored. One implant was lost after 5.5 years. The mean mandibular symphysis height was 8.9 ± 2.2, 16.4 ± 2.7, 15.7 ± 2.7, and 15.4 ± 2.5 mm at intake, after augmentation, after implantation, and at the last recall visit, respectively. Mean radiographic peri-implant bone loss was 0.6 ± 0.7 mm. Mean clinical index scores were very low. Patient satisfaction was high. Surgical complications related to the donor site were seroma ( n = 1), haematoma ( n = 2), and sensory disturbance of the lateral femoral cutaneous nerve ( n = 1); all had resolved before placement of the implants. Eleven patients reported postsurgical sensory disturbances of the mental nerve, of whom five still experienced some sensory disturbance at the last recall visit. Augmentation of the extremely resorbed mandible with an iliac crest onlay graft followed by placement of two implants 4 months later provides a solid basis for a bar-retained overdenture with favourable clinical and radiographic results.
Currently, dental implants serving as a basis for removable and fixed prostheses are considered a significant tool to reduce problems related to lack of retention of a mandibular denture. In many cases dental implants can be placed without much effort, but in the extremely resorbed mandible, the placement of dental implants still poses a challenge. It has been reported that in extremely resorbed edentulous mandibles, the placement of four 6-mm dental implants provides a solid basis for a bar-retained mandibular overdenture. In more extreme cases, reliable implant placement is not possible and an appropriate treatment is to augment the mandible prior to placement of endosseous implants. Common techniques to augment a mandible are onlay grafts and interpositional grafts. When applying an interposition technique, materials such as hydroxyapatite can serve as an alternative to autogenous materials (bone, cartilage), either in combination with autogenous bone or even as the only material applied. For onlay grafts, only autogenous bone has been used as a grafting material to date.
Depending on clinical circumstances, endosseous implants can be inserted during the same session as reconstruction of the resorbed mandible (simultaneous procedure), or after incorporation of the graft (delayed procedure, commonly after 3–4 months). A major advantage of the simultaneous procedure is that the graft and the implant can be incorporated at the same time, thus eliminating the need for a second operation. An important drawback of this approach is that the positioning and angulation of the implants is more complicated, making it less desirable from a prosthetic point of view. Another major drawback is the unpredictable resorption of the peri-implant bone in the grafted area. Verhoeven et al. showed that up to 50% of the grafted bone, including the area where the implants were placed, was resorbed within 10 years. Also, van der Meij et al. noted a clinically relevant resorption of the iliac crest onlay graft when applied simultaneously with the two implants (mean follow-up 4.3 years, 15% bone resorption). Therefore, it has been advised that implants be placed 4–6 months after grafting (delayed procedure), as then most of the loss of vertical bone height due to remodelling has already taken place and thus the extent of peri-implant bone loss will be significantly lower.
Hypothetically the mandible with a bone onlay graft will resorb to a level comparable to a mandible without implants and an onlay graft.
No studies have yet been published on the long-term treatment outcomes of onlay grafting of the severely resorbed mandible to facilitate implant placement. Therefore, a retrospective observational study was performed to assess the treatment outcomes of lower dentures on two endosteal implants placed in the severely atrophied mandible reconstructed with bone grafts from the iliac crest (delayed approach). The study was done in consecutive patients with a follow-up of at least 5 years.
Materials and methods
All consecutive patients treated between 2000 and 2007 with an iliac crest bone onlay graft and insertion of two implants in the extremely atrophied mandible, with a delayed procedure, were invited by letter to attend the clinic between September 2012 and November 2012 for a recall visit. Patients with a history of radiotherapy to the head and neck region, or a history of pre-prosthetic surgery or previous oral implantology, were excluded from the study.
Forty patients agreed to participate in the study (for details see section “Results”). The institutional review board approved the study. Written informed consent was obtained from all patients. The mean follow-up after fabrication of the prosthetic construction was 6.6 ± 1.2 years (range 5–9 years, median 6.0 years). Surgery and prosthodontics were performed within the same clinic.
Panoramic radiographs of one of the patients are shown in Fig. 1 . Fig. 2 illustrates the grafting procedure. All patients were treated by the same surgeon (GMR). Immediately preoperatively, broad-spectrum antibiotics (3 g amoxicillin) and 0.5 mg/kg dexamethasone were given intravenously. Under general anaesthesia, a block of corticocancellous bone was harvested, and some additional cancellous bone was taken in a standard fashion from the medial surface of the anterior ilium. Next to this, a low vestibular incision was made. A mucoperiosteal flap was raised towards the top of the alveolar ridge. After identifying the mental foramen on both sides, the flap was raised further to create a sub-periosteal tunnel distal to the mental foramen. Care was taken to avoid the mental nerve, and for this reason the tunnel was placed slightly to the lingual surface of the mandible. Subsequently, the iliac bone graft was cut with a saw. The resulting corticocancellous bone block was fitted as an interforaminal onlay graft and fixed with two screws (1.5-mm diameter; Martin Medizin Technik, Tuttlingen, Germany; Fig. 1 B) to the mandible ( Fig. 2 A). After grafting, the height of the mandible in the interforaminal area had to be at least 12 mm. The remaining cancellous bone was milled and used to fill the created sub-periosteal tunnel distal to the mental foramen. In this way, a smooth border was created between the grafted interforaminal area and the areas distal to the mental foramen ( Fig. 1 B). The mucoperiosteal flap was closed, after mobilization, in two layers with 3-0 polyglactin sutures. Postoperatively the patients were instructed to use 0.2% chlorhexidine mouth rinse (1 min, five times daily) for 2 weeks.
Four weeks after surgery, the dental prosthesis was corrected and the patient was allowed to wear it again. After an uncompromised healing period of 4 months, the grafted area was exposed under local anaesthesia and the screws used for fixation of the onlay graft were removed. Subsequently, two endosseous Straumann standard implants (Institute Straumann AG, Basel, Switzerland) with a diameter of 4.1 mm and a length of at least 10 mm were inserted in a one-stage procedure ( Fig. 1 C).
Two weeks after implant placement, the patient was allowed to wear the dentures again after relieving the denture in the area of the implants and relining it with a resilient material (Soft Liner; GC Corporation, Tokyo, Japan). The patient was given oral hygiene instructions. After a 3-month osseointegration period, a mandibular overdenture retained by a milled titanium bar ( Fig. 2 B), with gold retention clips attached to it, was made ( Fig. 1 D). All patients were on a recall schedule of at least once a year after treatment.
Outcome measures were implant survival, change in height of the mandible, change in the peri-implant bone level, condition of the soft tissue (plaque index, presence of calculus, gingival index, sulcus bleeding index, and pocket probing depth), and patient satisfaction. Panoramic radiographs were taken for all patients at the recall visit. The other radiographs and data on complications during surgery, postoperative healing (inflammation, wound dehiscence, sequestration and loss of bone particles, and sensory disturbances of the mental and lateral femoral cutaneous nerves ), and loss of implants were obtained from the patient records.
Loose and lost implants were scored at any time after placement.
Changes in height of the mandible
For all patients, standardized panoramic radiographs were taken as routine before surgery, after augmentation surgery, after placement of the implants, and at the last recall visit. The midline and the mental foramina were used as reference points. Vertical perpendicular lines were drawn through these points. The distance between the midline and the line through the mental foramen was divided into two equal parts and a line was drawn. In this way the change in height of the mandible was measured on these radiographs at five sites: at the midline (site 1), between the midline and mental foramen (sites 2 and 3), and at the mental foramen (sites 4 and 5). The height of the mandible was calculated at each site by measuring the distance between the points of intersection of the line and the inferior and superior borders of the mandible. The height was corrected with the known magnification factor of the panoramic radiograph.
Changes in peri-implant bone level
Peri-implant bone level changes were calculated using the panoramic radiographs. The known implant length was used as a reference to transform the linear measurements into millimetres. The reference line was the outer border of the neck of the implant. Mesial and distal bone changes in this region were considered as peri-implant bone changes and were defined as the difference in bone height between the radiograph taken after implant placement and the radiograph taken at the last recall visit, i.e., at least 5 years after placement of the prosthesis.
For the presence of plaque, the index of Mombelli et al. was used : score 0, no detection of plaque; score 1, plaque can be detected by running a probe across the smooth marginal surface of the abutment and implant; score 2, plaque can be seen with the naked eye; score 3, abundance of plaque. The presence of calculus was scored as 1, and the absence was scored as 0.
To qualify the degree of peri-implant inflammation, the modified Löe and Silness index was used: score 0, normal peri-implant mucosa; score 1, mild inflammation, slight change in colour, slight oedema; score 2, moderate inflammation, redness, oedema and glazing; score 3, severe inflammation, marked redness and oedema, ulceration.
For bleeding, the bleeding index of Mombelli et al. was used: score 0, no bleeding when using a periodontal probe; score 1, isolated bleeding spots visible; score 2, a confluent red line of blood along the mucosa margin; score 3, heavy or profuse bleeding.
The probing depth was measured at four sites of each implant: mesially, labially, distally, and lingually, using a periodontal probe (Merrit-B; Hu-Friedy, Chicago, IL, USA). The distance between the marginal border of the mucosa and the tip of the periodontal probe was scored as the probing depth.
Sensory changes of the skin were examined by stroking the upper lip and cheek with a cotton pellet and by pinching the lower lip with a pair of tweezers.
Patient satisfaction was assessed using a validated questionnaire consisting of items focusing on the upper and lower dentures separately, functional problem complaints in general, specific features related to facial and denture aesthetics, and accidental lip, cheek, and tongue biting. The extent of each specific complaint could be expressed on a four-point rating scale: 0 = no complaints; 1 = few complaints; 2 = moderate complaints; 3 = severe complaints.
The patient’s eating ability with their full denture was assessed using a chewing ability questionnaire. This questionnaire focused on how well the patient could eat soft, tough, and hard foods. There were three possible answers: 0 = good; 1 = moderate; 2 = bad.
Patient overall treatment satisfaction was expressed on a 10-point rating scale, with 1 = very bad, to 10 = excellent.
Data were collected by one observer (GCB); analysis of the radiographs was done by another observer (HJAM). The worst score per implant of the clinical and radiographic parameters was used in the data analysis.
The data were tested for normal distribution with a Kolmogorov–Smirnov test and were checked visually using a histogram with a distribution curve. As all data were normally distributed, differences between evaluation periods were tested with a paired Student’s t -test. A significance level of 0.05 was chosen for all tests. The scores of the questionnaire, the clinical parameters, and the overall satisfaction scores were not normally distributed. Therefore medians instead of means are given for these results. The analysis was done using IBM SPSS Statistics for Windows, version 20.0 (IBM Corp., Armonk, NY, USA).
Between 2000 and 2007, 48 patients were referred to the institution with an extremely resorbed mandible and persistent problems with the retention and function of their conventional mandibular dentures. They were treated with an implant-retained mandibular denture on two implants placed in the interforaminal region of the mandible, which had been reconstructed with an iliac crest onlay graft. The symphyseal height at baseline was 8.9 ± 2.2 mm (Cawood class VI , Fig. 1 A). Eight patients were excluded from the study: three had died and five had moved without informing the hospital of their new address. The remaining 40 patients (33 women and seven men; mean age 61 years, range 48–79 years at the time of surgery) agreed to participate in this study.
Early surgical complications related to bone grafting from the iliac crest were seroma ( n = 1), haematoma ( n = 2), and a sensory disturbance of the lateral femoral cutaneous nerve ( n = 1). All these complaints resolved completely during the remodelling phase of the grafted area, i.e., before insertion of the implants. The implants were inserted after 3.8 ± 0.6 months. Eleven patients reported postsurgical sensory disturbances of the mental nerve (objectively and subjectively), five of whom still had a sensory disturbance in the chin region at their last recall visit, but the size of this region had diminished over time. No signs of peri-mucositis or peri-implantitis were seen around the implants and no vestibuloplasty had to be performed during the follow-up period.
One implant was lost due to peri-implantitis 5.5 years after implant placement (5-year implant survival 98.7%). The patient who lost one implant received another implant; osseointegration of this replacement implant was uneventful and it functioned well.
Changes in height of the mandible
The mean height in the mandibular symphysis region was 8.9 ± 2.2, 16.4 ± 2.7, 15.7 ± 2.7, and 15.4 ± 2.5 mm at intake, after augmentation, after implantation, and at the last recall visit, respectively. When looking in detail at the change in bone height over time, a significant interforaminal increase in height of the mandible was observed at all five measuring points between baseline and augmentation ( P < 0.001), while a significant interforaminal decrease in height of the mandible was observed between augmentation and implant placement (for P -values see Table 1 ). Post implant placement, a significant decrease in interforaminal bone height was observed at two measuring points ( P = 0.021 and P = 0.016; Table 1 , Fig. 3 ).
|4: Right foramen||2: Right, between midline and foramen||1: Midline||3: Left, between midline and foramen||5: Left foramen|
|Baseline to augmentation||6.1 ± 3.0
( P < 0.001)
|7.5 ± 2.3
( P < 0.001)
|7.5 ± 2.3
( P < 0.001)
|7.2 ± 2.5
( P < 0.001)
|5.5 ± 3.1
( P < 0.001)
|Augmentation to implant placement||−1.2 ± 2.1
( P = 0.001)
|−1.3 ± 1.3
( P < 0.001)
|−0.7 ± 1.5
( P = 0.007)
|−0.6 ± 1.6
( P = 0.023)
|−1.2 ± 2.1
( P = 0.001)
|Implant placement to recall visit||−0.4 ± 1.6
( P = 0.109)
|−0.5 ± 1.3
( P = 0.021)
|−0.3 ± 1.3
( P = 0.216)
|−0.6 ± 1.4
( P = 0.016)
|−0.1 ± 1.5
( P = 0.801)