Secondary Logo

Journal Logo


Iliac crest donor site analgesia with ropivacaine infusion after autogenous bone graft

Bonvini, J. M.; Dullenkopf, A.; Borgeat, A.; Rentsch, K.

Author Information
European Journal of Anaesthesiology: March 2004 - Volume 21 - Issue 3 - p 239-241


Pain at the bone donor site is one of the major causes of morbidity after iliac crest bone graft and is often of greater significance to the patient than the pain from the surgery that required the autogenous bone graft [1]. We had a unique opportunity to assess the efficacy of postoperative analgesia after cancellous autogenous bone grafts from the iliac crest. A local infusion of ropivacaine 0.4% was used continuously and it was compared (in a double-blind fashion, in the same patient, at the same time and for the same procedure) with a local saline infusion and with morphine patient-controlled intravenous (i.v.) analgesia.

A 38-yr-old male (ASA II, 171 cm height, 71 kg body weight) was admitted to hospital for elective osteosynthesis after a traumatic left lower leg fracture. The patient's past clinical history was unremarkable, except for nicotine abuse of 30 pack-yr. Because of the size of the defect in the left tibia, autogenous cancellous bone was needed from both iliac crests. After institutional Ethics Committee approval and informed written consent were obtained, the patient was instructed on the day before surgery on pain assessment by a visual analogue score (VAS). The patient was also informed that he would be asked to give separate, independent scores for his pain in the right and the left iliac crests as well as for the left lower limb.

On the day of operation, the patient was premedicated with midazolam 7.5 mg orally 1 h preoperation. He was monitored with continuous three-lead electrocardiogram, non-invasive blood pressure and pulse oximetry. Intravenous access was obtained with an 18-G cannula in a forearm vein followed by infusion of Ringer's lactate solution 8 mL kg−1 h−1. Induction of general anaesthesia was performed using the target-controlled infusion technique with propofol. Fentanyl 0.2 mg and rocuronium 70 mg were given for analgesia and muscle relaxation, respectively. Propacetamol 2 g i.v. was administered before skin incision. Anaesthetic and surgical procedures were uneventful.

Cancellous bone was obtained from both iliac crests. Before closing the wounds a Spongostan® (Johnson and Johnson Ltd, Skipton, UK) dressing, measuring 3.5 × 2.5 cm, was placed in direct contact with the bone. Using an 18-G Tuohy needle, the skin was tunnelled laterally to the surgical incision so that a 20-G catheter could be placed in direct contact with the dressing. After the surgical procedure was completed, the catheters were then secured to the skin by sutures and adhesive dressing. Syringes for the initial bolus, as well as those for the continuous infusion, were prepared by hospital pharmacists and labelled A and B. Their content remained undisclosed to both the anaesthetists and patient until the end of the observation period. Before the conclusion of anaesthesia, a bolus of 30 mL from syringe A (ropivacaine 0.5%) on the left side and 30 mL from syringe B (NaCl) on the right side were injected through the corresponding catheters. The patient's trachea was then extubated and he was transferred to the intermediate care unit. Next, a continuous infusion of 8 mL h−1 was started, and continued for the next 48 h, on the left side with syringes from set A (ropivacaine 0.4%). On the right side, an infusion at the same rate was started with syringes from set B (NaCl). At the same time, patient-controlled i.v. morphine analgesia was arranged using a programmable pump (Pain Management Provider®; Abbott, Chicago, IL, USA) with the following settings: no basal infusion, 2 mg bolus and lock-out time 10 min. Supplemental propacetamol 2 g i.v. four times a day and rofecoxib 50 mg orally once a day were also administered.

On the patient's arrival at the intermediate care unit (time, t = 0) and every 8 h afterwards for 48 h - as well as 24 h after stopping the infusions (t = 72 h) - pain was registered by a trained, protocol-independent nurse using a VAS (range: 0 = no pain; 100 = worst pain imaginable) in the left and right hip as well as in the lower limb at rest.

Plasma ropivacaine and α-1-acid glycoprotein (AAG) concentrations were determined at t = 0 (arrival at the intermediate care unit, 60 min after bolus injection) and t = 48 h. The total ropivacaine plasma concentration was determined by gas chromatography using a nitrogen-sensitive detector; the free plasma ropivacaine fraction was determined by coupled-column liquid chromatography with mass spectrometric detection using electrospray ionization after ultrafiltration of the sample. Concentrations of AAG were measured by immunoturbidimetry. Statistical analysis of the pain scores between the two iliac crests was compared by paired t-tests. P < 0.05 was taken as being statistically significant. This case shows that continuous infusion of ropivacaine 0.4% at the iliac crest bone donor site provides significantly better pain control compared with placebo. Also remarkable is that there was a unique setting with the patient being under his own control.

Pain at the donor site caused patient discomfort for several months after the procedure, often requiring further treatment [1]. Todd and Reed [2] showed that infiltrating the wound with bupivacaine during operations for the removal of iliac bone grafts led to a significant reduction of pain, but only in the first 4 h postsurgery. Reuben and colleagues [3] showed that injecting morphine at the donor site also provided better analgesia compared with saline and morphine given intramuscularly; it also reduced chronic pain at the donor iliac crest. Further, Wilkes and Thomas [4] showed that continuous infusions of bupivacaine at the donor site effectively reduced pain 24 h after surgery compared with a 'single-shot' injection. These results are consistent with the present findings. In our patient, pain at the left iliac crest (ropivacaine infusion) was significantly less than pain at the right iliac crest (placebo infusion). Furthermore, at several time points, pain at the bone donor site with saline infusion was equal to or even higher than pain registered in the left lower limb, where the actual site of the defect requiring the surgery was located, despite the morphine patient-controlled anaesthesia (Fig. 1).

Figure 1
Figure 1:
Time-course of postoperative pain in the left iliac crest, ropivacaine side (IC Sx) (♦); in the right iliac crest, placebo side (IC Dx) (▴); and in the left shin (columns), assessed by a visual analogue scale (VAS). Time 0: arrival at the intermediate care unit. *P < 0.05.

During the first 8 h postoperation, the VAS score was 0 (no pain) on the left (ropivacaine) side. After this time, there was an increase in the VAS scores despite the continuous ropivacaine infusion. This could be a direct effect of the difference in ropivacaine concentration between the initial bolus (0.5%) and the continuous infusion (0.4%), as well as the slow disappearance of the effect of intraoperatively administered opioids.

Ropivacaine was chosen as local anaesthetic in this setting since it has less cardiac and central nervous system toxicity than bupivacaine [5]. Reabsorption of the drug from the iliac crest donor site follows different and unknown pharmacokinetics than from the epidural space, but even at the iliac crest, involuntary intravascular catheter placement or displacement is the most feared complication that could theoretically lead to potentially toxic plasma drug concentrations. Iliac crest catheters were placed under direct sight by the surgeon, and the absence of large vascular structures in the immediate vicinity means this unfortunate event is relatively unlikely to occur. Since the catheter was placed in the bed of the harvest site with many open bony interstices, it was embedded in the Spongostan® dressing to avoid close communication with the bony vascular structures.

In our patient, continuous infusion of ropivacaine was performed with a 0.4% solution at 8 mL h−1, which was effective during preliminary investigations. Despite this more concentrated ropivacaine solution, the measured concentrations at 48 h were still lower than values reported by Knudsen and colleagues [5] in healthy volunteers (Table 1), and no sign of local anaesthetic toxicity was noted. This could be partly explained by the increase in the degree of protein binding of ropivacaine. It is known that ropivacaine, as well as other local anaesthetics, mainly binds to AAG, whose blood concentrations rapidly increase in the first hour after surgery [6]. Total ropivacaine concentrations closely correlate with plasma levels of AAG [7], but the concentration of the unbound form, which is responsible for toxic effects, is mainly dependent on liver metabolism. An increase in AAG levels could buffer the increased total ropivacaine load, reducing its free fraction, therefore reducing its potential toxicity. These facts are consistent with the present results and could explain why at t = 48 h, despite a 5.4 times increase in the concentration of total ropivacaine, the free fraction was raised only by 1.5 times.

Table 1
Table 1:
Ropivacaine and α-1-glycoprotein concentrations.

Ropivacaine 0.4% infusion during the first 48 h postoperation at the iliac crest bone graft donor site significantly reduces pain without leading to toxic plasma concentrations in the present patient. However, large prospective studies are needed to assess the efficacy and safety of this technique, as well as to determine the optimal ropivacaine concentration in this setting and its impact on the chronic iliac crest donor site pain.

J. M. Bonvini

A. Dullenkopf

A. Borgeat

Department of Anaesthesiology; Balgrist University Hospital; Zurich, Switzerland

K. Rentsch

Department of Clinical Chemistry; University Hospital Zurich; Zurich, Switzerland


1. Hayes WR Jr, Smith RW. Trochanteric bone grafts in foot and ankle surgery. Foot Ankle Int 1996; 17: 402-405.
2. Todd BD, Reed SC. The use of bupivacaine to relieve pain at iliac graft donor sites. Int Orthop 1991; 15: 53-55.
3. Reuben SS, Vieira P, Faruqi S, Verghis A, Kilaru PA, Maciolek H. Local administration of morphine for analgesia after iliac bone graft harvest. Anesthesiology 2001; 95: 390-394.
4. Wilkes RA, Thomas WG. Bupivacaine infusion for iliac crest donor sites. J Bone Joint Surg Br 1994; 76: 503.
5. Knudsen K, Beckman SM, Blomberg S, Sjovall J, Edvardsson N. Central nervous and cardiovascular effects of i.v. infusions of ropivacaine, bupivacaine and placebo in volunteers. Br J Anaesth 1997; 78: 507-514.
6. Aronsen KF, Ekelund G, Kindmark CO, Laurell CB. Sequential changes of plasma proteins after surgical trauma. Scand J Clin Lab Invest Suppl 1972; 124: 127-136.
7. Erichsen CJ, Sjovall J, Kehlet H, Hedlund C, Arvidsson T. Pharmacokinetics and analgesic effect of ropivacaine during continuous epidural infusion for postoperative pain relief. Anesthesiology 1996; 84: 834-842.
© 2004 European Society of Anaesthesiology