Gentamicin distribution from a collagen carrier

Journal of Orthopaedic Research 14749-754 The Journal of Bone and Joint Surgery, Inc. 0 19% Orthopaedic Research Society

Gentamicin Distribution from a Collagen Carrier

Sanjay Mehta, J. Stewart Humphrey, "Daniel I. Schenkman, Anthony V. Seaber, and Thomas Parker Vail

Division of Orthopaedic Surgery, and *Division of Laboratory Animal Resources and Department of Pathology, Duke University Medical Center, Durham, North Carolina, U S A .

Summary: Local delivery of antibiotics by a degradable carrier has the potential for high local antibiotic levels and avoids systemic toxicity. Intravenous access, renal function monitoring, and subsequent surgical removal may not be required when degradable local delivery modalities are used. This study examined the in vivo elution of gentamicin from processed bovine collagen (type I) in 66 adult White rabbits. Collagen impregnated with gentamicin (3 mg/kg) was implanted into the vastus lateralis, and data were collected from 15 minutes to 28 days after implantation. Local tissue biopsies were taken a minimum of 2 mm from the implantation site. The gentamicin was released into the local tissue and averaged more than 3,800 pg/ml during the initial 4 hours after implantation. Local levels fell to 6.90 ? 5.22 pglml at 24 hours and subsequently were 2.70 2 1.75 pg/ml or more through day 28. Serum levels reached an average peak of 4.04 +- 1.75 pg/ml at 5 hours after implantation, decreased after the initial 24 hours, and subsequently were less than 0.41 ? 0.20 pg/ml through day 28. Collagen impregnated with gentamicin proved to be an effective degradable carrier of gentamicin in the healthy rabbit; it provided local tissue concentrations above the minimum inhibitory concentration and serum concentrations below levels associated with systemic toxicity for 28 days after implantation.

The treatment of acute bone infections, soft-tissue injuries, and osteomyelitis with intravenous antibiotics may produce secondary complications including nephrotoxicity, ototoxicity, and hypersensitivity (9,12, 14,28). As an alternative to intravenous therapy, Buch- holz and Gartmann introduced the concept of using polymethylmethacrylate cement mixed with antibiotics to achieve local tissue delivery in total joint arthroplasty (5) . Klemm (17) continued clinical investigation of the original idea of Buchholz and Gartmann, pion- eering the use of polymethylmethacrylate bead chains to treat orthopaedic infections. Others have provided evidence that local delivery by polymethylmethacrylate alone or concomitantly with systemic antibiotics is more effective than systemic therapy alone in treat- ing orthopaedic infections (13,14,16). A prospective clinical study by Salvati et al. (27) has shown that polymethylmethacrylate delivery can provide local concentrations 17 times higher than intravenous delivery while reducing both serum and urine concentrations 20 times. Canine modeling by Petty et al. (25) demonstrated statistically significant reductions in infection rates after total hip arthroplasty in dogs

Received October 25,1995; accepted April 25,1996. Address correspondence and reprint requests to T. P. Vail at

Duke University Medical Center, Box 3332, Durham, NC 27710, U.S.A.

treated with polymethylmethacrylate containing gentamicin as compared with other antimicrobial proto- cols including intravenous delivery. One drawback of polymethylmethacrylate beads as a delivery vehicle is that their use requires a second invasive procedure for removal (3,7,19). In addition, polymethylmethacrylate beads are typically prepared in the operating room (10); this increases operating room time and can intro- duce undesirable nonuniformity in the beads. Poly- methylmethacrylate has also been associated with mild adverse effects on the immune response and local microcirculation (21-24). There is also great variability in elution characteristics from polymethylmethacrylate among antibiotics (4). These variabilit- ies are contingent on the size and shape of the beads in addition to their antibiotic concentration (6).

An absorbable carrier of antibiotic would have the advantage of local delivery while eliminating the drawbacks of polymethylmethacrylate. Collagen impregnated with antibiotic does not require subsequent surgical removal, has been used successfully in Eu- rope, where it is commercially available, since 1987 (29), and carries the advantage of being a biodegrad- able carrier (30). Riegels-Nielsen et al. (26) have provided evidence that collagen with gentamicin is as effective in preventing infection as intravenous gentamicin in an animal model. Clinical studies comparing the effectiveness of gentamicin-impregnated collagen

749

75 0 S. MEHTA ET AL.

with that of gentamicin-loaded polymethylmethacrylate have demonstrated no statistically significant dif- ferences in outcomes (15,18).

The objective of this work was to examine the in vivo elution characteristics of gentamicin from locally implanted processed bovine collagen (type I). Specif- ically, local tissue gentamicin levels, serum gentamicin levels, and urine gentamicin levels were quantified and local histology was examined.

METHODS The methods in this study were approved by the Institutional

Animal Care and Use Committee.

Manufacture of CollagedGentamicin Implant The implant was prepared from bovine Achilles tendons, which

contain predominantly type-I collagen. The tendons were shred- ded and then were digested with pepsin according to the method described by Trelstad (32). The digested tendons were then separated from the undigested tendon fragments and centrifuged. After resuspension, the procedure was repeated several times to purify the implant material. One hundred milliliters of gentamicin solution, containing gentamicin (1.3 g) as sulfate (4.0 g), was added to 400 ml of a 0.8% acidic collagen solution; the solution was then pipetted into plastic trays. The final implant was prepared by freeze-drying these preparations into a 10 x 10 x 0.5 cm sponge containing 280 mg of collagen and 130 mg of gentamicin as sulfate. The implants were prepared and supplied by the manufacturer (Inocoll, Saal/Donau, Germany).

Experimental Protocol Collagen was implanted into the right lower extremity of 66

adult White rabbits. The rabbits were divided into two groups: group I (46 animals) received collagen impregnated with gentamicin and group I1 (20 animals) received only collagen. The vastus lateralis muscle was selected as the implantation site due to its large size and easy accessibility. For group I, the gentamicin- impregnated sponges were measured and cut to deliver a dose of 3 mg gentamicin per kilogram of body weight of the animal. These animals were killed and studied for local and systemic antibiotic levels as well as local histology at 6 hours (n = 5 ) . 12 hours (n = 5), 24 hours (n = 5), 36 hours (n = 5), 48 hours (n = 5), 5 days (n = 6), 7 days (n = 5), 10 days (n = 5), and 28 days (n = 5) after implantation. For group 11, plain collagen was implanted as a con- trol, using the same volume of material per kilogram of body weight, to study local histology. These animals were studied for local histology at 24 hours (n = S), 48 hours (n = 5), 7 days (n = 5 ) , and 28 days (n = 5) after implantation.

Group-I animals randomly assigned to the 6 hour study were anesthetized for the entire period. During the study period, an intravenous catheter was placed in the central ear vein and 2-3 cc of blood was collected every 15 minutes for the first hour and then every half hour until death. In this group, tissue samples were also collected every 2 hours from the site of implantation for assay of local gentamicin levels. Urine and tissue samples for histological examination were collected at death.

Operative Technique All operative procedures were performed on the right leg, and

aseptic technique was used. A lateral approach was employed. Skin, superficial fascia, and deep fascia were incised to approach the vastus lateralis muscle. The muscle was incised for 5-7 cm in length and 2 cm in depth. Stay sutures (3-0 nonabsorbable) were placed at equal intervals into the cut edges of the muscle and held by hemostats. The stay sutures were used to help retain collagen

Local Gent. Levels

in CI g/m I

-r

250.

200:

1.75 - L -

0.5 1 1.5 2 5 ? 10 28

Time in Days

F1G. 1. Average local gentamicin levels recorded in the first 6 hours after implantation were in the thousands of micrograms. The fall in local gentamicin levels that had begun at 6 hours after implantation continued sharply, as testified by the large negative slope from 0.5 to 1 day. After this fall, the values rose and then were subsequently maintained at about 10 pg/ml through the tenth day. Local levels tapered to a low of 2.7 2 1.75 pglml on day 28 after implantation.

inside the muscle. Suture (1-0 nonabsorbable) was used to decrease the inflammatory response that might be associated with dissolvable suture material. The calculated amount of collagen was cut into strips of equal length. Care was taken to place the strips symmetrically into the muscular defect. The implant geometry and biopsy site were standardized in all animals. All biopsies were performed at a distance of 2 mm from the stay sutures. The tensor fascia and skin were closed in layers.

The animals were anesthetized prior to death, and 2-3 cc of blood was collected to assay serum gentamicin levels. All animals were killed, at the given time intervals after implantation, with an intracardiac injection of 2-3 cc of sodium pentobarbital. Immedi- ately afterward, a spot urine sample was collected directly from the bladder, and tissue was taken for assay of local gentamicin levels and for histological examination. The histology samples were placed in 10% formalin for fixation. The time intervals were chosen on the basis of preliminary elution curves generated in a pilot study and work done by Grimm (11).

Biopsy Technique The distal one-third of the muscle implanted with collagen was

always selected for assay of local tissue gentamicin levels. After measuring the muscle and separating the distal one-third from the proximal two-thirds, a 2 mm section of muscle adjacent to the implant was discarded. The sample for assay was taken from the remaining muscle adjacent to the discarded edge. This was done to prevent a spuriously high local gentamicin level by the inadver- tent inclusion of implant in the biopsy sample. The tissue was placed in a sterile glass container and frozen. It was then forwarded to the Roche Laboratories (Research Triangle Park, NC, U.S.A.) where it was weighed, homogenized, and assayed for local gentamicin levels.

Gentamicin Assay Technique The TDx system (Abbott Laboratories, Abbott Park, IL,U.S.A.)

was used to assay gentamicin levels; it employs fluorescence po- larization immunoassay technology and is sensitive to 0.27 pgiml with 95% confidence (1). The technique is based on the principle that fluorescence of a fluorescein-labeled antimicrobial agent is quenched by antimicrobic specific antibodies. Thus, if there is a large amount of unlabeled antibiotic (gentamicin) in the sample following assay, the labeled antimicrobic will be displaced from the antibody and there is little quenching. From a standard curve

J Orthop Res, Vol. 14, No. 5, 1996

GENTAMICIN DISTRIBUTION FROM A C O L L A G E N CARRIER 75 I

4.5

4

3.5

Local 3 . - Gent. -.- Levels

in 2. - W m l 1.5.-

1

0.5

constructed with known amounts of a drug, values for the un- known can be determined (1). No physical separation of the bound and free fractions is required. The TDx system does not measure bioavailability or pharmacologic activity against microbes.

implant (Fig. 4), and by the tenth day no hemorrhage could be Seen in the muscle. ne collagen implant was grossly visible until the tenth day. By day 28, the implant could not be seen on gross examination.

- .- .-

.-

.. 0 , : : : : : : : : : : : : : t

RESULTS Gentamicin Levels

The peak concentration of gentamicin in local tissue at a minimum of 2 mm from the site of implantation was 3,848.97 ? 1,473.78 pg/ml (mean 5 SD) and occurred 4 hours after implantation of the gentamicin- impregnated collagen. The peak serum concentration was 4.04 -+ 1.75 pg/ml and occurred 5 hours after implantation. The maximum concentration in urine was 171.48 2 104.48 pg/ml and occurred 12 hours after implantation. In local tissue, the gentamicin levels recorded in the first 6 hours were in the thousands of micrograms. The average values in local tissue were 3,810.94 ? 1,991.46 pg/ml at 2 hours, 3,848.97 ? 1,473.78 pg/ml at 4 hours, and 1,543.30 * 763.73 pg/ml at 6 hours after implantation. The decrease in local tissue levels, which began at 6 hours after implantation, continued rapidly. After this sharp fall, the values rose and were then maintained at about 10 pg/ml through the tenth day. During the next 18 days, the local levels tapered off to a low of 2.7 ? 1.75 pg/ml (Fig. 1).

Gentamicin levels in serum were monitored at narrow intervals over the first 6 hours, as shown in Fig. 2. The peak level was 4.04 5 1.5 pg/ml, recorded at 5 hours after implantation (Fig. 2). Serum levels over the period from 6 hours to 28 days showed two features: a sharp decrease in concentration between 12 and 24 hours after implantation and a narrow range (0.24 5 0.08 pg/ml to 0.42 ? 0.15 pg/ml) within which the values were maintained over the next 27 days (Fig. 3).

The peak average value recorded from the spot urine assay was 171.48 2 104.48 pg/ml. The total volume of gentamicin excreted was not measured. Peak values in local tissue (4 hours after implantation) preceded peak values in serum ( 5 hours after implantation), which preceded peak values in urine (12 hours after implantation).

Morphology Gross Examination The muscle was hemorrhagic around the surgical

site initially. The extent and amount of hemorrhage grossly visible was greatest at 24 hours. This diminished over time, such that by the tenth day the external surface showed no visible hemorrhage. On the cut sections, hemorrhage was seen extending circumfer- entially around the implanted collagen. This hemorrhage also diminished over time, such that by the seventh day it extended for only 0.5-1 mm around the

Microscopic Examination The myofibers surrounding the implanted collagen

showed a loss of sharp corners and striations in the muscle fiber for a distance of 0.5-1 mm around the implant. The change was greatest in the superficial or external region of the muscle. This surrounding zone of histological change was identifiable as early as 6 hours after implantation. Over the next 4 days, the zone became established, and by the fifth day a clear line of demarcation had developed between the degenerating and normal muscle. Necrosis of the myofibers followed, as evidenced by the complete loss of shape by the myofibers and by extrusion of the nu- cleus. These necrotic myofibers were then replaced by fibroblasts, and the region became fibrotic. The fibrotic stage was seen in the longer groups (7,10, and 28 days).

r

Local Gent. Levels

in W m l

0 0!5 1 1.5 2 5 7 10 28

I

Time in Days

FIG. 3. The declining serum gentamicin concentrations first noted 6 hours after implantation continued through the 24 hours after implantation. From day 1 through day 28, serum concentrations were maintained within a narrow range (0.42 2 0.15 pglml to 0.24 2 0.08 pglml).

J Orthop Re& Vol. 14, No. 5, 1996

752 S. M E H T A ET AL.

FIG. 4. Histological section (hematoxylin and eosin) of rabbit muscle with gentamicin-impregnated collagen that had been implanted 7 days previously. A zone of myodegeneration with some inflammation is present around the collagen implant (A). Some blood has per- colated among the fibers of the implanted collagen (B).

Inflammatory Response In the group-I animals (gentamicin-impregnated

collagen), polymorphonuclear leukocytes were seen in the fascia1 planes as early as 6 hours after implantation. They were especially evident in and around the blood vessels. The inflammatory response inten- sified, as judged by increasing prevalence of acute inflammatory cells, and peaked by 24 hours. Also by

24 hours after implantation, polymorphonuclear leukocytes had surrounded the degenerating myofibers. By 48 hours, macrophages and fibroblasts were present in addition to the neutrophils. Degenerating neutrophils with pyknotic nuclei were present among the degenerating myofibers.

By the fifth day, the inflammatory infiltrate had shifted predominantly to chronic cellular elements, in-

FIG. 5. Histological section (hematoxylin and eosin) from rabbit muscle with gentamicin-impregnated collagen that had been implanted 7 days previously. Mineralized material (M) is separated by fibrous connective tissue (F), and multinucleated giant cells (G) are present- features of calcinosis circumscripta.


GENTAMICIN DISTRIBUTION FROM A COLLAGEN CARRIER 753

cluding monocytes and fibroblasts. Although the polymorphonuclear cells were still seen, they were less frequent. The main cellular constituents were fibroblasts and macrophages, which were seen in and around the degenerating myofibers. Foreign-body giant cells were also noted.

In group TI (plain collagen), the overall inflammatory response of the surrounding muscle was similar to the response observed in group I. Myofiber damage was less. The plain collagen implant maintained its histologic integrity until the fifth day, when it started to show vacuolation. The vacuoles were occupied by erythrocytes and polymorphonuclear leukocytes. The vacuolation increased over the next 7-10 days, and additional macrophages infiltrated the collagen. The degradation of collagen increased such that by day 28 no collagen was visible histologically at xl00 magnifi- cation. The musculature surrounding the implant site was repopulated with fibrous tissue.

Within the zone of degenerating muscle, calcifica- tion of myofibers, which became surrounded by giant cells and walled off by fibroblasts (features of calcinosis circumscripta) (Fig. 5 ) , was noted in several 7-day specimens from both group I and group 11.

DISCUSSION Our objective was to determine the in vivo elution

characteristics of gentamicin from locally implanted processed bovine collagen. With use of the collagen sponge, extremely high local tissue levels of gentamicin were achieved despite consistently low serum levels. The average peak concentration in urine followed the average peaks in local tissue and serum. Bioab- sorption was clearly evident histologically, along with inflammation and a zone of muscle degeneration.

The peak local tissue level was 3,848.97 2 1,473.78 pg/ml, recorded at 4 hours after implantation. At 2 hours after implantation, the average local level was 3,810.94 t 1,991.46 pg/ml. After the initial rapid release, the local level averaged 204.29 t 211.41 pg/ml 12 hours after implantation of the sponge and then decreased to an approximate average of 12 pg/ml over the initial 10 days. Between 10 and 28 days, the average local level decreased from 10.66 2 9.10 pg/ml to 2.70 5 1.75 pg/ml. The peak serum level was 4.04 2 1.75 pg/ml, recorded 5 hours after implantation of the sponge. Subsequently the serum level began a gradual decline, remaining less than 0.42 t 0.15 pg/ml after 24 hours. According to Ascherl et al. (2), the delivery of antibiotic from the collagen matrix is dictated initially by washout and secondarily by breakup of mo- lecular affinities. This may explain the second peak, 3,810.94 2 1,991.46 pg/ml, in local tissue recorded over the first 6 hours.

Gentamicin eluted from the collagen into adjacent tissue and then into the systemic circulation, resulting

in a 2 hour delay between peak local tissue concentration and peak serum concentration. The delay in reaching the systemic circulation also accounts for the delay between the peak gentamicin concentration in local tissue and the peak concentration in urine, recorded 12 hours after implantation of the gentamicin- impregnated collagen sponge. The large SDs for values in urine likely were due to the variable concentration of urine in the bladder at death, the hydration status of the rabbit, and possibly intrinsic individual metabo- lic variations among the rabbits.

The implant clearly demonstrated bioabsorption histologically. The implant was grossly visible until the tenth day. By day 28, it could not be seen. Micro- scopic examination revealed a 1-2 mm zone of degenerating muscle around the implant, evident at 6 hours after implantation and clearly demarcated on day 5 after implantation. This degenerative zone was most prominent deep to the implant and less prominent superficial to the implant. This led us to hypothesize that the degenerative zone resulted, at least in part, from surgical trauma and pressure exerted by the implant onto adjacent tissue. An inflammatory response was evident at 6 hours after implantation and peaked at 24 hours.

Efficiency and a lower risk of gentamicin toxicity could be achieved with high local levels and low serum levels of the drug. Sustained peak serum concentrations of gentamicin greater than 12-15 pg/ml for longer than 10 days are associated with increased risk of nephrotoxicity and ototoxicity (31). Sustained trough serum concentrations of more than 2 pg/ml are associated with increased risk of toxicity (31). The peak serum concentration is associated with improved therapeutic success when gentamicin is given intrave- nously. The therapeutic concentration for gentamicin was defined by Noone et al. (20) as being greater than 5 pg/ml for patients with soft-tissue infections. Intra- venous delivery (8) can result in much higher peak serum concentrations (30-32 pg/ml) to achieve desired local levels compared with local delivery (4 pg/ml). With delivery of gentamicin by the intravenous route, the serum levels peaked immediately at introduction of the bolus dose and decreased over the next 6 hours to baseline levels of 1 pg/ml. With local delivery by collagen, the serum levels peaked 5 hours after introduction into the body and took 24 hours to reach the baseline serum level of 0.5 pg/ml. The baseline serum level of 0.5 pg/ml is below the trough levels reported to be associated with toxicity (0.5-2 pg/ml).

Collagen impregnated with gentamicin, as examined in this animal model, was found to deliver nontoxic systemic levels while providing high local concentrations in the first 24 hours and concentrations in the minimum inhibitory concentration range for as long as 28 days. This study supports continued experimen-


754 S. MEHTA ET A L .

tation with local delivery of antibiotic to address the issue of bioavailablity and efficacy in an infection model.

Acknowledgment: The authors gratefully recognize the support of Inocoll, SaaVDonau, Germany, for supplying the collagen/gentamicin implants used in this study and Roche Laboratories, Re- search Triangle Park, NC, U.S.A., for performing the gentamicin assays.

REFERENCES 1. Abbott Laboratories: TDxFlx@TDx@ Assays Manual. Abbott

Park, Illinois, Abbott Laboratories, 1994 2. Ascherl R , Stemberger A, Geissdorter K, Claudi B, Machka

K, Lechner F, Blumel G: Experimental results and preliminary clinical experience of collagen with gentamicin as an adjuvant measure in chronic osteomyelitis. In: Collagen as a Drug Carrier: Some Applications in Surgery, pp 34-40. Ed by A Stemberger, R Ascherl, F Lechner, and G Bliimel. Munich, Franklin Scientific, 1991

3. Becker PL, Smith RA, Williams RS, Dutkowsky JP: Compar- ison of antibiotic release from polymethylmethacrylate beads and sponge collagen. J Orthop Res 12737-741,1994

4. Brien WW, Salvati EA, Klein R, Brause B, Stern S: Antibiotic impregnated bone cement in total hip arthroplasty: an in vivo comparison of the elution properties of tobramycin and van- comycin. Clin Orthop 296:242-248, 1993

5. Buchholz HW, Gartmann HD: Infektionsprophylaxe und operative Behandlung der schleichenden tiefen Infektion bei der totalen Endoprothese. Chirurg 43:446-453,1972

6. Calhoun JH, Mader J T Antibiotic beads in the management of surgical infections. A m J Surg 157:443-449, 1989

7 . Chang CC, Merritt K. Effect of Staphylococcus epidermidis on adherence of Pseudomonas aeruginosa and Proteus mira- bilis to polymethyl methacrylate (PMMA) and gentamicin- containing PMMA. J Orrhop Res 9:284-288,1991

8. Curl JL, Curl JS: Pharmacokinetics of gentamicin in laboratory rabbits. A m J Vet Res 49:2065-2067,1988

9. Federspil P, Schatzle W, Tiesler E: Pharmacokinetics and ototoxicity of gentamicin, tobramycin, and amikacin. J Infect Dis 134(Suppl):S200-S205,1976

10. Fish DN, Hoffman HM, Danziger LH: Antibiotic-impregnated cement use in U.S. hospitals. A m J Hosp Pharm 49:2469-2474, 1992

11. Grimm H: Bacteriological and pharmacokinetic aspects of locally applied antibiotics. In: Collagen as a Drug Carrier: Some Applications in Surgery, pp 27-30. Ed by A Stemberger, R Ascherl, F Lechner, and G Blumel. Munich, Franklin Sci- entific, 1991

12. Hau T: [Antibiotic therapy in surgery: limitations and errors.] Zentribl Chir 1161379-1389,1991

13. Henry SL, Ostermann PAW, Seligson D: The prophylactic use of antibiotic impregnated beads in open fractures. J Trauma

14. Henry SL, Seligson D, Mangino P, Popham GJ: Antibiotic- impregnated beads. Part I: Bead implantation versus systemic therapy. Orthop Rev 20:242-247,1991

30:1231-1238,1990

15. Ipsen T, Jorgensen PS, Damholt V, Torholm C: Gentamicin- collagen sponge for local applications: 10 cases of chronic osteomyelitis followed for 1 year. Acta Orthop Scand 62592- 594,1991

16. Josefsson G, Lindberg L, Wiklander B: Systemic antibiotics and gentamicin-containing bone cement in the prophylaxis of postoperative infections in total hip arthroplasty. Clin Or-

17. Klemm KW: Gentamicin-PMMA chains (Septopal chains) for the local antibiotic treatment of chronic osteomyelitis. Reconstr Surg Traumatol20:11-35, 1988

18. Letsch R, Rosenthal E, Joka T [Local antibiotic administra- tion in osteomyelitis treatment: a comparative study with two different carrier substances.] Aktuel Traumatol 23:324-329, 1993

19. Miclau T, Dahners LE, Lindsey R W In vitro pharmacokinetics of antibiotic release from locally implantable materials. J Orthop Res 11:627-632,1993

20. Noone P, Parsons TMC, Pattison JR, Slack RD, Garfield- Davies D, Hughes K: Experience in monitoring gentamicin therapy during treatment of serious gram-negative sepsis. Br Med J 1:477-481,1974

21. Panush RS, Petty R W Inhibition of human lymphocyte re- sponses by methylmethacrylate. Clin Orthop 134:356-363,1978

22. Petty R W The effect of methylmethacrylate on the bacterial inhibiting properties of normal human serum. Trans Orthop Res SOC 2:137,1977

23. Petty R W The effect of methylmethacrylate on chemotaxis of human polymorphonuclear leukocytes. Trans Orthop Res Soc 2:138,1977

24. Petty W, Caldwell JR: The effect of methylmethacrylate on complement activity. Clin Orthop 128:354-359, 1977

25. Petty W, Spanier S, Shuster JJ: Prevention of infection after total joint replacement: experiments with a canine model. J Bone Joint Surg [Am] 70536-539,1988

26. Riegels-Nielsen P, Espersen F, Holmich LR, Frimodt-Moller N: Collagen with gentamicin for prophylaxis of postoperative infection: Staphylococcus aureus osteomyelitis studied in rabbits. Acta Orthop Scand 66:69-72, 1995

27. Salvati EA, Callaghan JJ, Brause BD, Klein RJ, Small RD: Reimplantation in infection: elution of gentamicin from cement and beads. Clin Orthop 207:83-93,1986

28. Sande MA, Mandell GL: Antimicrobial agents: the aminogly- cosides. In: Goodman and Gilman’s The Pharmacological Ba- sis of Therapeutics, pp 1098-1116. Ed by AG Gilman,TW Rall, AS Nies, and P Taylor. New York, Pergamon Press, 1990

29. Steinig HJ: Collagen with gentamicin in the treatment of osteomyelitis. In: Collagen as a Drug Carrier: Some Applica- tions in Surgery, pp 53-56. Ed by A Stemberger, R Ascherl, F Lechner, and G Blumel. Munich, Franklin Scientific, 1991

30. Stemberger A, Ascherl R, Lechner F, Blumel G: Collagen as a Drug Carrier: Some Applications in Surgery. Ed by A Stem- berger, R Ascherl, F Lechner, and G Blumel. Munich, Frank- lin Scientific, 1991

31. Taylor WJ, Caviness MHD [eds]: A Textbook for the Clinical Application o f Therapeutic Drug Monitoring, p 294. Irving, Texas, Abbott Laboratories, Diagnostic Division, 1986

32. Trelstad RL: Native collagen fractionation. In: Zmrnunochem- istry of the Extracellular Matrix, vol 1, pp 32-39. Ed by H Furthmayr. Boca Raton, CRC Press, 1982

thop 1591194-200,1981


Documents

Gentamicin distribution from a collagen carrier