Prevention of heterotopic ossification after spinal cord injury with COX-2 selective inhibitor (rofecoxib).

STUDY DESIGN: A randomized, prospective, double-blind, placebo-controlled clinical trial. OBJECTIVES: To determine the effect of COX-2-selective inhibitor on the prevention of heterotopic ossification (HO) after spinal cord injury (SCI). SETTING: County and University Teaching Hospital, Miami, FL, USA. METHODS: A total of 76 patients were enrolled in the study. Among them, 39 patients received placebo, and 37 received COX-2-selective inhibitor rofecoxib 25 mg daily for a period of 4 weeks. Prevention was started 3 weeks after spinal cord injury (SCI). In both groups of patients there was similar age as well as the level of SCI and ASIA impairment scale. Two methods were used to diagnose early HO, clinical symptoms and bone scintigraphy. Radiography was used for diagnosis of late stages of HO development. RESULTS: A significantly lower incidence of HO was found in the rofecoxib group (13.4%) than in the placebo group (33.3%: P<0.05). In patients receiving rofecoxib, there was a 2.5 times lower relative risk of developing HO than in the placebo group (95% CI, 2.3-6). There were no patients who discontinued the study due to adverse effects of medication. CONCLUSION: Our data suggest that COX-2-selective inhibitor rofecoxib is an effective medication in prevention of HO after SCI.

Prevention of heterotopic ossification after spinal cord injury with indomethacin.

STUDY DESIGN: A randomized, prospective, double-blind, placebo-controlled clinical trial. OBJECTIVES: To determine the effect of indomethacin on the prevention of heterotopic ossification (HO) following spinal cord injury (SCI). SETTING: County Hospital, Miami, Florida, USA. METHODS: Sixteen patients were treated with slow-release indomethacin 75 mg daily and 17 patients received placebo for a period of 3 weeks. Prevention was started 21+/-14 days after SCI. In both groups of patients there was similar age of the patients as well as the level of SCI and ASIA impairment scale. Two methods were used to diagnose HO, bone scintigraphy and radiographic examination. Bone scintigraphy with technetium labeled methylene-diphosphonate was used for diagnosis of early stage, while radiography was used for diagnosis of late stage of HO development. RESULTS: A significantly lower incidence of early HO was found in the indomethacin group (25%) than in the placebo group (65%; P<0.001). Similarly there was a significant reduction of late HO in the indomethacin group (12.5%) as compared to the placebo group (41%; P<0.001). CONCLUSION: Our data suggest that indomethacin used during the first 2 months after SCI is effective in prevention of HO in a significant number of patients.

The effect of etidronate on late development of heterotopic ossification after spinal cord injury.

Forty patients with spinal cord injury (SCI) and heterotopic ossification (HO) were treated with etidronate and followed after therapy to determine the effects of long-term medication on heterotopic bone formation. Eighteen patients had tetraplegia and 22 had paraplegia. Early diagnosis of HO (positive bone scintigraphy and negative radiographic findings of HO) was established by 3-phase bone scintigraphy using 99m technetium-labeled methylene diphosphonate. All patients underwent treatment with etidronate, first with intravenous administration of 300 mg/day for 3 days followed by an oral administration of 20 mg/kg/day for 6 months. Eleven patients (27.5%) developed radiographic evidence of HO from 1.5 to 6 years after therapy. A low degree of HO was found in these patients; 8 had grade I and 3 had grade II ectopic ossification (Brooker's scale). The analysis of data showed that 2 different types of ectopic bone may form in the later stages after SCI. In 5% of patients, HO was found in the same anatomical site initially and finally, suggesting a "rebound" in mineralization of bone matrix not prevented by the administration of etidronate. The other type of HO was found in the majority of patients (95%) where the localization of HO showed different involvement of joints than initially, indicating de novo appearance of HO following SCI. The data suggest that etidronate given for a prolonged period in higher doses has, in addition to an inhibitory effect on crystal formation, a cellular effect on bone-forming cells.

Triiodothyronine stimulates the release of membrane-bound alkaline phosphatase in osteoblastic cells.

Thyroid hormone deficient osteoblastic cells in cell culture released a significantly higher amount of alkaline phosphate (ALP) activity following T3 replacement. T3 increased the release of total and membrane-bound ALP activity in these cells significantly more than T4 or inactive thyroid hormone metabolite, DIT. The effect of T3 on the membrane-bound ALP fraction was dose and time dependent; higher concentrations of T3 and longer incubation time with T3 proportionally increased the enzyme activity. T3 had no effect on the release of soluble fraction of ALP. Our results indicate that in "hypothyroid" osteoblastic cells the total release of ALP is decreased and that the secreted fraction of ALP is predominantly in soluble form, whereas the addition of T3 stimulates ALP release and mainly increases the membrane-bound fraction. T3 also increased formation of actin cytoskeleton in hypothyroid osteoblastic cells. Cytochalasin treatment, through its inhibition of actin polymerization, produced a significant decrease of membrane-bound ALP release induced by T3. These data suggest that the regulatory role of T3 in skeletal development can partly be due to its stimulatory effect on the release of membrane-bound ALP by osteoblastic cells which is thought to be an important factor in the initiation of biological calcification.

Evaluation and management of heterotopic ossification in patients with spinal cord injury.

Sixty-three patients with paralysis secondary to spinal cord injury (SCI) were screened for heterotopic ossification (HO) by bone scintigraphy 27 +/- 14 (mean +/- SD) days after SCI. There were four female and 59 male patients, 36 had paraplegia and 27 tetraplegia. The age of patients was 28 +/- 9 years. Bone scintigraphy was obtained with a 3-phase test using 99m-technetium labeled diphosphonate, and the positive third phase was used as a criterion for diagnosis of HO. Bone scintigraphy was negative for HO in 27 patients (14 tetraplegic and 13 paraplegic) and positive in 36 patients (13 tetraplegic and 23 paraplegic). The patients with positive HO were treated with etidronate, first with an intravenous dose of 300 mg/day for 3 days, and then with an oral dose of 20 mg/kg/day for 6 months. The follow-up of patients consisted of clinical and radiographic evaluations every 2-4 months. The extent of HO was classified radiographically. In the treated group of patients who completed the entire course of etidronate therapy one patient developed HO, the remaining 28 (97%) patients had no radiographic evidence of HO during the follow-up of 10.6 +/- 4.5 months after initiation of therapy. Our data indicate that: (a) early HO can be detected in the asymptomatic patients using bone scintigraphy on the average of 4 weeks (27 +/- 14 days) after SCI and (b) the therapy with etidronate might be effective in the prevention of HO in majority of patients when the treatment is initiated in an early stage of heterotopic bone formation.

Treatment of heterotopic ossification after spinal cord injury.

A new protocol in management of heterotopic ossification (HO) was evaluated in 46 patients after spinal cord injury (SCI). A group of 24 paraplegic and 22 tetraplegic patients was involved in a prospective study. Diagnosis of HO was made by bone scintigraphy and radiographic evaluation. Patients were divided into two groups. Group I was made up of 33 patients with positive bone scintigraphy and negative evidence of HO and Group II was made up of 13 patients with positive bone scintigraphy and positive radiographic evidence of HO. Etidronate was started intravenously (300 mg/day) for three days followed by oral therapy for six months (20 mg/kg/day). Follow-up of patients was 15.7 +/- 8 months after SCI. In Group I, etidronate therapy prevented the development of HO in 79 percent of patients; in 21 percent of patients, a low degree of tissue ossification was found which was not clinically significant. In Group II, there was an inhibitory effect of etidronate on progression of soft tissue ossification in six patients. The remaining seven patients did not respond to therapy and showed an increased growth of HO. Our data indicate that etidronate may prevent HO in the majority of patients when administered at an early stage of HO development and in higher doses than are routinely recommended.

A study comparing sterile and nonsterile urethral catheterization in patients with spinal cord injury.

This study was designed to determine the effect of sterile and nonsterile intermittent catheterization on the incidence of urinary tract infection (UTI) in patients after spinal cord injury. The study included 29 patients with neurogenic bladder dysfunction treated with intermittent catheterization. One group of 14 patients was on sterile catheterization; another group of 15 patients was on nonsterile catheterization. On a weekly basis, urine samples were obtained and analyzed. A total of 122 urine samples were analyzed. The patients on sterile catheterization had a 28.6% UTI incidence; the group using a nonsterile catheterization technique had a UTI incidence of 42.4%. The most common urinary pathogen in both groups was E. coli (65%). The cost of antibiotics for patients on the sterile catheterization program was only 43% of the cost of antibiotics for those on the nonsterile program. However, the sterile kits cost 371% of the cost of the catheterization kits for the patients in the nonsterile program, so the total cost of managing neurogenic bladder on the sterile program was 277% of the cost of the nonsterile program.

Interaction of cultured rat osteoblasts with 8-methoxypsoralen during irradiation with long-wave ultraviolet light.

Rat osteoblasts in monolayer cell cultures have been irradiated with long-wave ultraviolet light (UVA) in the presence and without 8-methoxypsoralen (8-MOP). In the absence of 8-MOP, the exposures to UVA (3 x 10(-3)W.cm-2) for up to 30 min have not affected cellular viability, the rate of 14C-acetate incorporation, and alkaline phosphatase (AP) activity. However, it depressed 3H-TdR incorporation rates by osteoblasts. In the presence of 15 to 100ng of 8-MOP/ml, even 5-min irradiation of osteoblasts was sufficient to reduce DNA synthesis. Much higher (0.5 to 1.0 micrograms/ml) 8-MOP concentrations were required to depress lipid synthesis, AP activity, and the viability of irradiated cells. These results suggest that in osteoblasts the machinery of DNA synthesis is especially labile to photosensitization with 8-MOP and UVA, whereas UVA light by itself exerts a less potent inhibitory effect.

Increased muscle strength in paralyzed patients after spinal cord injury: effect of beta-2 adrenergic agonist.

The administration of beta-2 adrenergic agonists in experimental animals result in an increased strength of skeletal muscle. In this study, we evaluated whether a beta-2 adrenergic agonist, metaproterenol, had an effect on muscle size and strength in a group of patients with muscular atrophy following spinal cord injury. Ten male subjects were randomly divided into 2 groups and agreed to participate in a prospective, double-blind, placebo-controlled, and crossover study. Metaproterenol (80 mg/day), or placebo, was administered orally for a period of 4 weeks. Muscle strength was measured by a force transducer interfaced with a microcomputer. Muscle size was calculated and expressed as a cross-sectional area of upper arm and forearm using a formula. Metaproterenol induced a significant increase of muscle strength in both groups of subjects, compared with placebo (p < .001). Similarly, there was an increase in a muscle size in the forearm following the administration of metaproterenol. Our data indicate that beta-2 adrenergic agonists may improve both muscle strength and size in patients with muscular atrophy following spinal cord paralysis.

Effect of nonsteroidal antiinflammatory drugs on fracture healing: a laboratory study in rats.

We studied the effects of two nonsteroidal antiinflammatory drugs (NSAIDs) on fracture healing in rats: ibuprofen (30 mg/kg/day) and indomethacin (1 mg/kg/day). Femoral fractures were induced via a three-point bending technique. NSAIDs were administered orally for 4 or 12 weeks. Control animals received no medication. In each group a minimum of six animals were killed at the following intervals: 2, 4, 6, 8, 10, and 12 weeks postfracture. Fracture healing was determined by mechanical testing and histologic evaluation. The bending strength of each fractured femur was expressed as a percentage of the strength of the intact, contralateral femur. Histologic evaluation was performed on serial longitudinal sections stained with hematoxylin and eosin using a qualitative score of maturity of the callus. Ibuprofen and indomethacin both retarded fracture healing, with significant differences in "mechanical healing" found between the control and experimental groups after 10 weeks of drug administration. Both drugs also induced qualitative histologic changes manifested by delayed maturation of callus, which was noticeable earlier than the difference found by mechanical testing of bone. Our data suggest that NSAIDs have an inhibitory effect on fracture repair that is reversible after cessation of indomethacin but not ibuprofen.

Fracture healing and mast cells.

We analyzed the morphology and localization of mast cells during the course of fracture repair in control rats and in animals with delayed healing of fractures induced by nonsteroidal antiinflammatory drugs (NSAIDs). In the first 2 weeks of fracture healing in control animals, mast cells were found either in the vicinity of blood vessels or in the vascularized tissue proliferating into the cartilaginous portion of subperiosteal callus. In the later stages (6-8 weeks), mast cells were seen in loose connective tissue in bone marrow surrounded with translucent ground substance. At this stage of healing, a hyperplasia of mast cells and cell degranulation was often seen in close proximity to osteoclasts and areas of bone resorption. Treatment with NSAIDs delayed fracture healing and the appearance of mast cell hyperplasia in bone marrow for approximately 4 weeks, suggesting that mast cells have specifically defined temporal and regional distribution during the process of bone repair. Histochemical studies documented a significant amount of chymase in the mast cells in callus. This enzyme was purified from mast cells by chromatography and was able to digest in vitro proteins extracted from bone. Our data suggest that mast cells in fracture healing are involved in digestion of extracellular matrix in callus tissue that could facilitate (a) angiogenesis in the early stages of healing, and (b) the replacement of provisional tissue with newly formed bone in the later stages of fracture healing.

Gentamicin penetration into normal rabbit nucleus pulposus.

STUDY DESIGN: Radioactively labeled gentamicin was administered to 24 rabbits to assess the concentration of antibiotic in the nucleus pulposus. OBJECTIVES: The purpose of the study was to investigate the pharmacokinetics of gentamicin penetration into normal rabbit nucleus pulposus. SUMMARY OF BACKGROUND DATA: Disc space infection is a complication of spinal surgery that can be prevented by prophylactic antibiotics. Gentamicin can be used in conjunction with other antibiotics as a prophylactic agent. One previous study demonstrated that a similar antibiotic, tobramycin, penetrates the disc, but no data have been reported on the pharmacokinetics of disc penetration. METHODS: Twenty-four rabbits were given an intravenous injection of gentamicin labeled with iodine 125. Four rabbits were killed at hourly intervals 1 to 6 hours after injection. Specimens of nucleus pulposus, blood, whole liver, and saline-perfused liver were obtained and prepared. The radioactivity in the specimens was measured. RESULTS: The gentamicin concentration in the nucleus pulposus peaked at 2 hours and remained at this level for the duration of the experiment. Twenty percent of the gentamicin recovered from the nucleus pulposus was tissue bound. CONCLUSIONS: Gentamicin concentration in the rabbit nucleus pulposus does not peak until 2 hours after an intravenous bolus of drug. If gentamicin penetrates human nucleus pulposus in a similar fashion, this study could have implications for the timing of administration of this agent for prophylaxis.

Tissue distribution of antibiotics in the intervertebral disc.

STUDY DESIGN: This study analyzed the distribution of antibiotics within the intervertebral disc of rabbits. Specimens were tested with specific antibodies against antibiotics using an immunofluorescent technique. OBJECTIVES: The results were correlated to provide a rationale for perioperative prophylaxis of infection. SUMMARY OF BACKGROUND DATA: Several groups of investigators and the recent data from our laboratory showed quantitative changes in penetration of antibiotics into intervertebral disc. No previous study has assessed antibiotic distribution in anulus fibrosus and nucleus pulposus. METHODS: Discs were obtained from rabbits after intravenous injection of penicillin or gentamicin. Antibiotics were localized in tissue sections using specific antibodies with a immunofluorescent method. RESULTS: Penicillin (negatively charged) and gentamicin (positively charged) penetrated the neutrally charged anulus fibrosus, but penicillin had less ability than gentamicin to penetrate into the negatively charged nucleus pulposus. CONCLUSION: Our data suggest that penetration and distribution of antibiotics into avascular intervertebral disc is significantly influenced by the charge of antibiotics.

Evaluation of serum osteoblast mitogenic activity in spinal cord and head injury patients with acute heterotopic ossification.

STUDY DESIGN: This was a blind, prospective study of the effect of sera from patients with spinal cord and head injuries on osteoblast proliferation. OBJECTIVES: The authors studied whether a humoral factor that stimulates the formation of heterotopic bone is released into the circulation after a neural injury. BACKGROUND DATA: Other authors have shown that a humoral osteoinductive factor may be released after head and spinal cord injuries. METHODS: Serum was obtained at certain times throughout the first 12 weeks post-injury and from control subjects. It was incubated with osteoblasts harvested from fetal rats, as well as with fibroblast controls. RESULTS: There was a significant rise in serum mitogenic activity after injury in both groups. When patients that developed heterotopic ossification were compared to other patients and controls, no significant differences were seen. CONCLUSIONS: This in vitro study fails to support a humoral mechanism for heterotopic ossification after spinal cord or brain injuries.

Morphologic and metabolic changes in rat osteoblast cultures during the dark reaction with 8-methoxypsoralen.

The dark reaction of 8-methoxypsoralen (8-MOP) with cultured rat osteoblasts did not cause significant changes in cellular replication rates or in the synthesis of RNA and proteins. Microscopic examination, however, revealed that the dark reaction resulted in massive accumulation of perinuclear lipids and in the statistically significant enhancement of alkaline phosphatase activity. A sharp, and statistically significant, upsurge of lipid synthesis in osteoblasts preceded microscopically detectable accumulation of lipids and occurred only during the initial, but not during the subsequent stages of the dark reaction. These results suggest that in the course of the dark reaction the plasma membrane of osteoblasts is a target of psoralen.

Penetration of glycopeptide antibiotics in nucleus pulposus.

The penetration of the glycopeptide antibiotics vancomycin and teicoplanin into the nucleus pulposus of rabbits was studied. Blood samples were obtained at 0.5, 1, 4, 8, and 24 hours after intravenous administration of 30 mg/kg vancomycin or 16 mg/kg teicoplanin. Concentrations of antibiotics were determined in tissue specimens and serum samples by fluorescence polarization immunoassays. Antimicrobial activity in the nucleus pulposus was determined with an agar diffusion method using a strain of Micrococcus luteus as the indicator organism. A peak concentration of vancomycin in the nucleus pulposus was attained 8 hours after drug administration. Teicoplanin reached its maximum level and plateaued 1 and 2 hours, respectively, after injection, and it remained unchanged for the rest of the study. This microbiologic analysis showed that the nucleus pulposus contained an antimicrobial level of glycopeptide antibiotics after administration. Because rabbit nucleus pulposus is similar anatomically to that of humans, these results may have implications regarding the timing and choice of antibiotic administration.

Intravenous disodium etidronate therapy in spinal cord injury patients with heterotopic ossification.

The goal of the present study was to use intravenous etidronate in the acute phase of heterotopic ossification (HO) in an attempt to achieve a high initial drug concentration at the site of the active ectopic ossification. The study included 27 consecutive patients with an acute onset of HO after spinal cord injury (SCI). The three-phase bone scan was used to confirm clinical diagnosis of HO. Disodium etidronate (Didronel) 300 mg was administered intravenously daily for 3 to 5 days. In 20 patients there was a rapid (1-2 days) decrease of soft tissue swelling (p < 0.01) with no side effects associated with the intravenous administration. In seven patients there was minimal or no improvement of edema after intravenous etidronate. In these patients deep vein thrombosis was found in the affected limbs. The effect of high dose etidronate on HO was determined in the group of 13 patients with positive clinical and scintigraphic finding of an acute HO, but negative radiographic studies. After intravenous administration of etidronate for 3 days (300 mg/day) the drug was continued orally with 20 mg/kg/day for 6 months. A placebo was not used in this study. In eight patients there was no radiographic evidence of HO after therapy while two patients had minimal ossifications. In three patients therapy was interrupted and all developed HO in 1-2 months.

Interaction of osteoblasts with extracellular matrix: effect of mast cell chymase.

We studied the effect of mast cell chymase on the interaction between osteoblasts and extracellular matrix. Chymase was purified from mast cell lysate by anion exchange chromatography. Osteoblasts were isolated from rat calvarias by collagenase digestion. Incubation of osteoblasts with mast cell lysate (40-170 micrograms/ml) or purified chymase (8-32 micrograms/ml) resulted in changes in cell-matrix interaction and cell morphology. Osteoblasts treated with chymase also showed a gradual detachment from the artificial substrata and from the biomatrix (collagen-digested rib fragment). A similar effect of mast cell chymase on the osteoblasts was found in vitro on endosteum of an intact parietal bone. Neutral protease inhibitors abolished the effect of both crude and purified enzyme preparations on the cell-matrix interaction. Mast cell chymase had no effect on osteoblast viability. The effect of enzyme on osteoblast proliferation was studied with lower concentrations of enzyme (0.2 micrograms/ml) in order to avoid cell detachment; there was no effect on either the metaphase index or on the number of cells after 5 days of incubation with chymase. Osteoblast attachment and cell spreading on different matrix proteins (fibronectin, vitronectin, extract of noncollagenous matrix proteins from rat bone) were significantly altered by their pretreatment with chymase. Matrix fibronectin of osteoblasts in culture as well as soluble vitronectin and fibronectin were digested by rat mast cell chymase. Our data suggest that mast cells through action of neutral protease chymase may alter molecules in extracellular matrix that are important in osteoblast adhesion, cell spreading, maintenance of cell morphology, and, most likely, cell function.

Effect of mast cell chymase on the morphology of thyroid cells in vitro.

We studied the effect of mast cell chymase on the thyroid cells in culture. Rat serosal mast cells, similar functionally to connective tissue mast cells, were obtained after lavage of the peritoneal cavity and lyzed by freezing. The resulting lysate was used as crude enzyme preparation. Mast cell chymase was purified from the crude preparation by anion exchange chromatography. Crude and purified chymase incubated with thyroid cells induced cellular retraction, the appearance of long processes and gradual cell detachment from the substratum. The effect of the enzyme was not cytotoxic. The immunofluorescence studies of thyroid cells showed a decreased amount of polymerized actin and tubulin after incubation with chymase. Neutral protease inhibitor abolished the effect of crude and purified chymase on thyroid cell morphology. The above findings suggest that mast cell chymase may have a function in the control of cell morphology and cell-matrix interaction.

The effect of mast cell chymase on extracellular matrix: studies in autoimmune thyroiditis and in cultured thyroid cells.

In the first part of the study we analyzed the morphology of mast cells in autoimmune thyroiditis of BB/W rats. In the early stage of thyroiditis mast cells showed exocytosis of granules into the interstitium; this was associated with disorganization of the extracellular matrix and the appearance of a translucent ground substance in stroma. Mast cells were not seen in the mononuclear infiltrates in the later stages of thyroiditis. In order to further study the effect of mast cells on the extracellular matrix, we evaluated the effect of mast cell lysate and purified chymase on the matrix of cultured thyroid cells. Mast cells were obtained from peritoneal cavity; mast cell chymase was purified by anion exchange chromatography. After exposure to chymase there was a reduction of pericellular fibronectin in cultured thyroid cells, while laminin in matrix remained unchanged. Similarly, as found by gel electrophoresis, soluble fibronectin and vitronectin were digested by chymase in the reaction mixture. Cell attachment on both fibronectin and vitronectin was significantly decreased upon exposure of matrix proteins to chymase. The effects of chymase were abolished by enzyme inhibitor phenylmethane sulfonyl fluoride. These data suggest that mast cells possess proteolytic enzymes capable of digesting different host proteins which may have a role in the thyroid cell interaction with the surrounding matrix.