Transplants from the anterior iliac crest are used for most reconstructive procedures in cranio-maxillofacial surgery. The advantages are easy accessibility, the ability to work in two teams and the amount of corticocancellous bone available; disadvantages are postoperative pain and gait disturbances. To reduce donor-site pain, the effect of a single dose of bupivacaine (10 cc of 2.5 mg/cc with 1:80.000 epinephrine) was studied. 200 consecutive patients, who underwent anterior iliac crest bone harvesting for reconstructive procedures, were randomly divided into those receiving bupivacaine and those not. They completed a standardized questionnaire. Patients scored the intensity of the pain and difficulties walking at different times with a visual analogue scale. They recorded analgesics used. 98 questionnaires were eligible for analysis. No differences between the bupivacaine and the control group were detected for postoperative pain and gait disturbance. There is no support for administration of a single dose of bupivacaine to reduce pain in the first postoperative days. The surface area of the removed bone had a significant influence on pain and walking; pain is related to the local osseous damage or periosteal stripping rather than to the length of incision or the operation time.
In oral and maxillofacial surgery, the anterior iliac crest is the most frequently used donor site for autogenous bone grafts, that are used in reconstructive procedures such as sinus floor elevation, augmentation procedures, closure of alveolar clefts and in post-traumatic or post-ablative reconstructive surgery. This site provides abundant corticocancellous bone and allows a two-team approach, but it is associated with acute or chronic pain, gait disturbance, infection, poor aesthetic results or neurologic injury .
Strategies to reduce postoperative donor-site pain include the administration of a single dose of a local anaesthetic at the donor site , application of epidural anesthesia , repeated bolus or continuous infusion of a local anaesthetic via an indwelling catheter or a change in operative technique . To decrease postoperative pain, at the end of the bone harvesting procedure, the authors routinely infiltrate 10 cc of bupivacaine (2.5 mg/cc with 1:80.000 epinephrine) into the donor site. There is little evidence in the literature to support this regimen. To test the hypothesis that a single-dose of bupivacaine reduces short-term and/or long-term donor-site pain, a prospective randomized controlled clinical trial was performed.
Material and methods
This prospective study comprised 200 consecutive patients, who underwent anterior iliac crest bone harvesting for reconstructive procedures in the period 2006–2007. All patients underwent surgery at the Department of Oral and Maxillofacial Surgery, Rijnstate Hospital, Arnhem, the Netherlands. Patients with conditions that might bias the postoperative pain scores, such as rheumatic disorders, low back pain, coxarthrosis, prosthesis of the hip or knees or previous iliac crest surgery, were excluded. After obtaining informed consent, for each patient an envelope was drawn, thereby enrolling the patients randomly into group A which received bupivacaine or into the control group (group B) which did not receive bupivacaine. Patients were not informed in which group they were included.
Early effects of harvesting iliac crest bone (<4 weeks postoperative)
Patients were asked to fill out a questionnaire. Using a visual analogue scale (pain-VAS; range 0–100 mm) patients scored the intensity of the pain at three moments on the day of surgery (t 0a , t 0b , t 0c ), and the first postoperative day (t 1a , t 1b , t 1c ) and subsequently at days 2 (t 2 ), 3 (t 3 ), 4 (t 4 ), 5 (t 5 ), 6 (t 6 ), 7 (t 7 ), 14 (t 14 ), 21 (t 21 ) and 28 (t 28 ). At t 1 , t 7 , t 14 , t 21 and t 28 patients were also asked to note any difficulties during walking (walking-VAS; range 0–100 mm). Patients were instructed to keep a strict record of the analgesics (paracetamol/ibuprofen) they used at all times (t 0-28 ). They were instructed only to score the pain that was specifically associated with the donor site. At t 0 and t 1 the average scores for the pain-VAS were calculated for the three measurements.
During surgery the following surgical parameters were measured: length of the incision, duration of the bone harvesting procedure (incision until wound closure), length and width of the removed bone, and blood loss. Multiplying length and width of the harvested bone resulted in the surface area of the donor region. Demographic data were obtained such as age, sex, weight, length. By dividing weight in kilograms by height in meters squared, the body mass index (BMI) was calculated.
Late effects of harvesting iliac crest bone (>1 year postoperative)
To assess long-term morbidity, pain, and to obtain additional information about patient satisfaction, a second questionnaire was sent to the respondents more than 1 year postoperatively.
Senior residents performed all bone-harvesting procedures. Patients were placed in a supine position with a hip roll placed to expose the anterior superior iliac spine (ASIS). While stretching the skin in a superomedial direction, an incision was made 1.5 cm posterior to the ASIS to reduce the risk of injury to the lateral cutaneous femoral nerve. Dissection through the subcutaneous tissue was towards the musculotendinous aponeurosis of the fascia lata muscle and oblique abdominal muscles. Taking care not to transect any muscle fibres, this aponeurosis was incised over the iliac crest with an anterior perpendicular releasing incision, maintaining a distance of at least 1.5 cm from the ASIS, to reduce the risk of fracture. The iliac muscle, its periosteum included, was then reflected to the medial side, exposing the medial cortex. Perpendicular to the crest, two vertical and two horizontal connecting bone-cuts were made using osteotomes. The superior horizontal bone-cut was made at the midcrestal level, allowing harvesting of a mono-cortical cortico-cancellous bone graft. Additional cancellous bone chips were harvested with curettes. A hemostatic material (Spongostan ® ) was applied at the donor site. The wound was closed in layers. First the periosteum was carefully sutured using Vicryl ® 2-0. The subcutaneous tissue was approximated by Vicryl ® 3-0. The cutis was closed using an intracutaneous resorbable monofilament suturing material (Monocryl ® ). No drain was left.
After wound closure, only patients in group A received a single dose of bupivacaine (10 cc of 2.5 mg/ml bupivacaine with 1:80.000 epinephrine), given as a depot, injected at the medial side of the iliac crest and on top of it.
Immediate postoperative pain management comprised intravenous administration of an opiate analgesic drug (10 mg morphine) and 75 mg diclofenac, a non-steroidal anti-inflammatory drug (NSAID). Patients were encouraged to ambulate the day after surgery. On discharge, paracetamol (1000 mg every 6 h) and a NSAID (ibuprofen 600 mg every 8 h) were prescribed.
For data analysis the statistical analysis software package SPSS 16.0.01 was used. An independent sample t-test was applied to assess differences between groups A and B in relation to the postoperative pain-VAS-score, walking-VAS-score, the total daily use of analgesic drugs and other assessed variables. A multivariate regression analysis of the combined groups A and B was performed to detect the influences of the various variables (BMI, age, sex, surface area of the donor site, incision length, duration of the bone harvesting procedure and the appliance of bupivacaine) on pain-VAS and walking-VAS.
Early postoperative effects of harvesting iliac crest bone (within 4 weeks)
Of the 130 questionnaires returned, 32 were excluded because they were not correctly filled-in, resulting in 98 questionnaires for evaluation (49%). For the control group B there were 42 questionnaires compared with 56 for group A.
Pre-implant surgery comprised mandibular augmentation (n = 14), maxillary sinus floor elevation procedures (n = 5) and buccal cortical onlay augmentation of the maxilla (n = 4). In 71 patients, both sinus floor elevation procedures, as well as onlay augmentation were performed. In 4 patients, in the same session both the maxilla and mandible were augmented.
During surgery, an unintended perforation of the lateral cortex of the iliac crest was observed in groups B (n = 4) and A (n = 5). No postoperative infections occurred. In one patient a fracture of the ASIS occurred, another developed a hematoma; both were treated conservatively. For all except one patient (n = 97), the postoperative hospital stay lasted only 1 day. This patient, with a BMI of 40.2, stayed for 11 days owing to significant problems during postoperative ambulation. An incision of 94 mm was needed to harvest the iliac bone graft in this patient. Table 1 gives an overview of the measured variables. Except for the BMI index, groups A and B were statistically comparable.
|Group A (range)||Group B (range)||Significance ( p < 0.05)|
|Age (years)||56 (21-74)||57 (21-80)|
|Length of incision (mm)||45 (30-65)||45 (28-94)|
|Surface area (cm 2 )||11.2 (3-27)||10.8 (2-31)|
|BMI-index (Kg/m 2 )||27.2 (19.3-42.5)||24.9 (16.8-40.2)||p = 0.025|
|Blood loss (cc)||85.8 (10-600)||115.3(5-750)|
|Operation time (min)||31.1 (17-52)||31.8 (15-68)|
Since the control group and the experimental group differ with respect to the BMI, the effect of the experimental variables on the outcome variables was analysed with a multivariate model, containing the use of bupivacaine and BMI as independent variables. Results from this analysis can be found in Table 2 . As can be seen, for all points in time and for both the walking-VAS and pain-VAS, bupivacaine had no statistically significant effect.
|p -value||Effect||p -value||Effect|
|Pain-VAS t 0||0,53||−3,7||0,03||1,2|
|Pain-VAS t 1||0,77||1,7||0,05||1,1|
|Pain-VAS t 2||0,31||5,7||0,01||1,4|
|Pain-VAS t 3||0,58||3||0,15||0,8|
|Pain-VAS t 4||0,99||0,1||0,25||0,6|
|Pain-VAS t 5||0,47||−3,9||0,23||0,6|
|Pain-VAS t 6||0,44||−4||0,15||0,7|
|Pain-VAS t 7||0,43||−4,1||0,23||0,6|
|Pain-VAS t 14||0,77||−1,5||0,16||0,7|
|Pain-VAS t 21||0,55||−2,5||0,20||0,5|
|Pain-VAS t 28||0,09||−5,9||0,01||0,9|
|Walking-VAS t 1||0,20||7,9||0,00||1,8|
|Walking-VAS t 7||0,78||−1,6||0,06||1,1|
|Walking-VAS t 14||0,79||−1,6||0,12||0,9|
|Walking-VAS t 21||0,71||−2||0,11||0,9|
|Walking-VAS t 28||0,52||−3,4||0,03||1,1|
In group A, the pain-VAS decreased from 36.3 (t 0 ) to 8.9 (t 28 ) with a peak of 41.2 at t 1 . In group B, the pain-VAS decreased from 36.8 (t 0 ) to 12.2 (t 28 ) with a peak of 37.2 at t 1 ( Fig. 1 ). With respect to the walking-VAS, the same tendency was observed for both group, a decrease from 52.2 (t 1 ) to 14.2 (t 28 ) in group A and from 40.9 (t 1 ) to 14.5 (t 28 ) in group B ( Fig. 2 ). No statistical significant differences were found between the groups.
An overall higher use of paracetamol was observed in group B, but it was only statistically significant at t 4 , t 5 , t 6 and t 7 ( Fig. 3 ). The total daily intake of ibuprofen at t 4 , t 5 , t 6 and t 7 seemed higher in group A, although it was not statistically significant ( Fig. 4 ). As expected, a gradual decrease in analgesic use was observed.