点击显示 收起
1 From the Department of Gastroenterology, Copenhagen University Hospital, Rigshospitalet, Copenhagen (KVH, LT, and MS), and the Department of Endocrinology, Hvidovre Hospital, Hvidovre, Denmark (HAS).
2 Address reprint requests to KV Haderslev, Department of Gastroenterology CA 2121, Copenhagen University Hospital, Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark. E-mail: khaderslev{at}dadlnet.dk. 3 Leiras Research, Finland, provided the study medication.
ABSTRACT
Background: Patients receiving home parenteral nutrition (HPN) because of intestinal failure are at high risk of developing osteoporosis.
Objective: We studied the effect of the bisphosphonate clodronate on bone mineral density (BMD) and markers of bone turnover in HPN patients.
Design: A 12-mo, double-blind, randomized, placebo-controlled trial was conducted to study the effect of 1500 mg clodronate, given intravenously every 3 mo for 1 y, in 20 HPN patients with a bone mass T score of the hip or lumbar spine of less than -1. The main outcome measure was the difference in the mean percentage change in the BMD of the lumbar spine measured by dual-energy X-ray absorptiometry. Secondary outcome measures included changes in the BMD of the hip, forearm, and total body and biochemical markers of bone turnover, ie, serum osteocalcin, urinary pyridinoline, and urinary deoxypyridinoline.
Results: The mean (±SEM) BMD of the lumbar spine increased by 0.8 ± 2.0% in the clodronate group and decreased by 1.6 ± 2.0% in the placebo group (P = 0.43). At all secondary skeletal sites (ie, hip, total body, and distal forearm), we observed no changes or small increases in the BMD of the clodronate group and decreases in the BMD of the placebo group. In the clodronate group, biochemical markers of bone resorption decreased significantly (P < 0.05).
Conclusions: Clodronate significantly inhibits bone resorption as assessed by changes in biochemical markers of bone turnover. Although the mean BMD increased in the clodronate group, cyclic clodronate therapy failed to increase spinal BMD significantly at 12 mo.
Key Words: Clodronate • home parenteral nutrition • bone resorption • osteoporosis • bone mass • bone turnover • bone mineral density
INTRODUCTION
Long-term parenteral nutrition self-administered in the home (HPN) is a lifesaving procedure in patients who cannot maintain adequate nutrition with enteral feeding alone because of intestinal failure. Most patients with nonmalignant diseases requiring HPN have Crohn disease, ischemic bowel infarction, radiation enteritis, or abnormal gut motility. Several studies have documented that patients receiving HPN frequently have osteoporosis (1, 2), and longitudinal studies have shown a continued loss of bone mineral during HPN treatment (2–4). Osteoporosis can cause substantial morbidity as a result of fractures and immobilization and may negatively affect the quality of life of these patients.
Bisphosphonates are drugs that selectively inhibit osteoclast-mediated bone resorption. These drugs are widely used in disorders associated with increased bone turnover and have become the mainstay for preventing and treating postmenopausal osteoporosis. Although the mechanism underlying the disturbed bone metabolism in HPN patients is not clearly defined, intervention with an agent that slows bone resorption may be a useful treatment option. However, the effect of bisphosphonate therapy on bone mineral density (BMD) and bone metabolism in HPN patients has not been examined in a controlled study.
Clodronate, a second-generation bisphosphonate, has long been used in the treatment of high-turnover bone diseases, particularly Paget disease and hypercalcemia of malignancy (5). Clodronate has also proven efficient for the treatment of postmenopausal osteoporosis, and both continuous and cyclical regimens given orally or intravenously have resulted in significant increases in BMD (6–13). In addition, several studies have shown that the long-term use of clodronate is associated with a decrease in fracture frequency in patients with neoplastic bone disease (14–17) and in postmenopausal women (11).
Bisphosphonates are most often administered orally, but unlike other bisphosphonates such as alendronate and risedronate, clodronate can be given both orally and intravenously. Because of the poor intestinal absorption of bisphosphonates in general, intravenous administration seems most appropriate in patients with intestinal failure. To evaluate the efficacy of intravenous clodronate therapy in preventing bone loss in HPN patients with low bone mass, we conducted a prospective 12-mo, randomized, double-blind, placebo-controlled study and also examined the effects of clodronate on biochemical markers of bone turnover in HPN patients.
SUBJECTS AND METHODS
Study participants
Ambulatory patients receiving HPN therapy for >1 y were eligible for entry if they had a BMD > 1.0 SD below the mean peak bone mass of either the lumbar spine or the femoral neck (determined by dual-energy X-ray absorptiometry; DXA). The exclusion criteria included renal insufficiency (a creatinine clearance rate <35 mL/min), evidence of disorders known to be associated with disturbances of bone and mineral metabolism, consumption of 3 alcoholic drinks/d, pregnancy or lactation, and current treatment with glucocorticoids or other drugs that would interfere with bone metabolism. Patients were also excluded if their spinal radiographs showed abnormalities that precluded accurate measurements with DXA. None of the patients were bedridden.
Clinical assessment at the time of entry included diagnosis, history of previous abdominal surgery (including the length of the remaining small intestine), duration of HPN therapy, composition of the parenteral nutrition supplied, body mass index [BMI; in kg (wt)/m2 (ht)], smoking habits, and menopausal status.
A total of 59 patients were evaluated for possible entry: 24 patients met the inclusion criteria and 20 patients agreed to participate and were enrolled. The diagnoses were as follows: Crohn disease (n = 10), ischemic bowel infarction (n = 2), systemic scleroderma (n = 1), volvulus (n = 1), chronic intestinal pseudoobstruction (n = 1), motility disorders (n = 2), intestinal pneumatosis (n = 1), Hodgkin lymphoma (n = 1), and intestinal lymphangiectasis (n = 1).
The study protocol was approved by the Ethics Committee for Medical Research in Copenhagen, and the study was conducted in accordance with the Declaration of Helsinki of 1975, as revised in 1983. Informed consent was obtained from all patients before entry.
Treatment and randomization
Patients were randomly assigned in consecutive order (in allocation blocks of 2) to receive either placebo or 1500 mg clodronate (Bonefos; Leiras, Oy, Turku, Finland) intravenously every 3 mo for 1 y. The vials of clodronate and placebo were identical in appearance. The information concerning the randomization was kept confidential in sealed envelopes until the end of the trial. Clodronate (1500 mg) was dissolved in 1500 mL isotonic sodium chloride solution and administered intravenously over 6 h. Before the infusion began, measures were taken to ensure that the patients were optimally hydrated, and no other intravenous fluids were given 2 h before, during, or 2 h after the infusion. To avoid fluid overload, the patient’s usual HPN dose was decreased by 1500 mL on the day of the study. During the study, the patients’ usual HPN program and medications were maintained. The details of composition and administration of HPN in our center were previously described (4). All patients received vitamin D supplements, which were provided routinely as a multivitamin in the parenteral solution at a dose of 400 IU ergocalciferol once weekly; in 3 patients, 100000 IU cholecalciferol was given intramuscularly once a month.
After entry, each patient was seen every 3 mo. At each visit, we conducted an interim history that included questions about possible adverse events. Clinical laboratory evaluations included analyses of serum and urinary biochemical indexes and of routine hematologic indexes every 3 mo. Although patients who did not adhere to the protocol were censored for the per-protocol analysis, every effort was made to further monitor these patients according to the trial schedule.
Measurement of bone mineral density
The BMD of the posterior-anterior spine, hip, distal forearm, and total body was measured at baseline and at 12 mo by DXA with a Norland XR-36 DXA densitometer (3.9.4 host software and 3.0 scanner software; Norland Corporation, Fort Atkinson, WI) according to the manufacturer’s instructions. BMD was also expressed as an SD score (z score) and as a T score. The T score was calculated according to the formula
RESULTS
Baseline data, dropouts, and adverse events
A total of 20 patients were enrolled, of whom 45% had osteopenia and 55% had osteoporosis: 10 were randomly assigned to receive clodronate therapy and 10 to receive placebo. One patient in the clodronate group died during the study period as a result of sepsis that was unrelated to the study protocol. Thus, data from 19 patients were available for the intention-to-treat analysis. One patient was deemed a protocol violator (did not receive the infusions at 6 and 9 mo); consequently, 18 patients (90%) remained for the per-protocol analysis. As shown in Table 1, the baseline demographics (including BMD measurements at all sites and calcium-regulating hormones) were not significantly different between the 2 study groups. Furthermore, there were no significant differences in the pretreatment biochemical variables of bone turnover between the 2 groups (P > 0.52 for all 3 markers). In the pooled study population, concentrations of osteocalcin, pyridinoline, and deoxypyridinoline were 1.7 ± 0.2 nmol/L, 77.4 ± 11.8 mmol/mmol creatinine, and 18.5 ± 3.18 mmol/mmol creatinine. These pretreatment values were all in the upper end or above the normal range, indicating high bone turnover at entry. These bone markers were not significantly different between the men and the pre- and postmenopausal women; mean values in the men and the premenopausal women were slightly greater than those in the postmenopausal women (P > 0.67). Additionally, we found significant positive correlations (r > 0.61, P < 0.01) between all 3 markers of bone turnover at onset.
View this table:
TABLE 1 . Baseline characteristics of the study patients1
The clodronate infusions were generally well tolerated by the patients, and no complaints of fever or myalgias were made. The only possible side effect reported was a tension headache, which occurred in 2 patients in the clodronate group and in 1 patient in the placebo group on one occasion. None of the infusions were stopped because of these events.
Changes in bone mineral density
In both study groups, the longitudinal changes in BMD were heterogeneous, with some subjects showing deterioration, some showing improvement, and some showing no changes. The individual results and the comparison of treatment groups when analyzed on a per-protocol basis are detailed in Table 2. The mean BMD of the lumbar spine decreased by 1.6 ± 2.0% in the placebo group and increased slightly by 0.8 ± 2.0% in the clodronate group (mean difference: 2.4%; P = 0.43) (Figure 1). The BMD of the hip decreased by 1.8 ± 2.2% in the placebo group and increased by 1.6 ± 3.0% in the clodronate group (mean difference: 3.4%; P = 0.36). The BMD of the distal forearm decreased by 3.2 ± 1.4% in the placebo group and increased significantly by 2.6 ± 1.3% in clodronate group (mean difference: 5.7%; P = 0.009). Total-body BMD decreased by an average of 2.9 ± 2.4% in the placebo group; in contrast, total-body BMD decreased only slightly (by 0.1 ± 1.5%) in the clodronate group. The results of the intention-to-treat analysis did not differ significantly from those of the per-protocol analysis. In both study groups, differences between the pre- and posttreatment BMD measurements were not significant, except for the difference in BMD of the distal forearm, which decreased in the placebo group (mean difference: 3.2%; P = 0.047).
View this table:
TABLE 2 . Percentage changes in bone mass by dual-energy X-ray absorptiometry in home-parenteral-nutrition patients after 12 mo of treatment with placebo or clodronate1
FIGURE 1. . Mean (±SEM) percentage changes in bone mineral density (BMD) in 18 home-parenteral-nutrition patients (per-protocol analysis) after 12 mo of treatment with clodronate () or placebo (). *Significantly different from the placebo group, P < 0.05 (Student’s t test).
For exploratory purposes only, a subgroup analysis was performed in the per-protocol population to disclose a possible relation between sex and the effect of treatment. In the women (n = 10), clodronate treatment consistently increased BMD compared with the placebo: by 6.3% (P = 0.09), 11.4% (P = 0.041), 9.5% (P = 0.01), and 0.8% (P = 0.64) in the spine, hip, arm, and total body, respectively. In the men (n = 8), clodronate treatment increased BMD in the arm by 1.0% (P = 0.74) and in the total body by 5.6% (P = 0.45) but decreased BMD in the spine by 2.2% (P = 0.67) and in the hip by 6.0% (P = 0.18). Slightly more postmenopausal women were allocated to the placebo group, which could have affected the results. However, exclusion of the postmenopausal women from the analysis did not change the overall results.
Fractures
Forty percent of the patients in the placebo group and 30% in the clodronate group had vertebral fractures at baseline. Three new vertebral fractures were noted during the study in 2 patients in the placebo group, whereas no new vertebral fractures were noted in the clodronate group (P = 0.47). Only one nonvertebral fracture occurred during the study period, which was a hip fracture in a patient in the placebo group.
Biochemical markers of bone turnover and routine laboratory tests
Treatment with clodronate was associated with a significant change in markers of bone turnover compared with placebo (P < 0.05 for all markers), and the P value for the interaction between treatment and time was <0.1 for all markers. The reduction in the bone-resorption markers urinary pyridinoline and urinary deoxypyridinoline occurred within 3 mo of the initiation of clodronate treatment and became significant after 6 mo (Figure 2). The reduction was sustained or slightly accentuated throughout the 12-mo period. By the end of the study, pyridinoline and deoxypyridinoline concentrations had decreased by 32 ± 7% (P = 0.07) and 34 ± 8% (P < 0.01), respectively. Correspondingly, the bone-formation marker serum osteocalcin sharply increased initially, with a significant peak at 3 mo (51 ± 16%; P < 0.01) that was followed by a gradual decrease toward baseline. We found strong positive correlations between changes in the 2 markers of bone resorption at 3, 6, and 12 mo (r > 0.8, P < 0.05 for all correlations) but no significant correlations between changes in osteocalcin and changes in pyridinoline and deoxypyridinoline. Also, there were no significant correlations between changes in any of these 3 bone markers and changes in BMD measurements at 12 mo in the clodronate group.
FIGURE 2. . Mean (±SEM) percentage changes from baseline in markers of bone turnover in 18 home-parenteral-nutrition patients receiving clodronate (•) or placebo () for 12 mo. *Significantly different from the placebo group, P < 0.05 (ANOVA).
To explore the influence of the baseline rate of bone turnover on the response to clodronate, patients in the clodronate group were categorized into 3 subgroups: low, normal, or high biochemical indexes of bone turnover. For each patient, the somewhat arbitrary total effect on BMD was calculated as the average value of the change in BMD of the spine, hip, forearm, and total body at 12 mo The mean total effect on BMD was 2.2%, 1.9%, and -2.8% in the patients with a low, normal, or high rate of bone turnover at baseline, respectively.
The results of the routine laboratory blood tests (eg, measurements of hematologic indexes, liver biochemical indexes, acute phase proteins, electrolytes, creatinine, and alkaline phosphatase) showed no systematic deviations between the 2 study groups. However, clodronate treatment resulted in a mild, transient, but significant decrease in ionized calcium and phosphate after a treatment period of 3 mo.
DISCUSSION
This study showed that treatment of HPN patients with 1500 mg clodronate intravenously every 3 mo was safe and associated with a significant inhibition of bone resorption, as assessed by markers of bone turnover. However, active treatment with clodronate did not increase the BMD of the lumbar spine more so than did the placebo. Clodronate treatment resulted in a nonsignificant 2.4% greater increase in the mean BMD of the spine relative to placebo, and a beneficial trend was observed at all secondary skeletal sites. Compared with the placebo, clodronate induced a significant increase in the BMD of the distal forearm in the group as a whole and a significant increase in the BMD of the hip and distal forearm in the female subgroup. A decrease in BMD was noted in the placebo group, with an average loss of spinal, hip, and arm BMD of 1–3%. This annual loss of BMD is slightly less than that previously reported in HPN patients (3, 4), possibly because the patients in the current study had no active disease or other concurrent disorders that could interfere with bone metabolism.
The optimal regimen of clodronate administration for the treatment of postmenopausal osteoporosis is not yet established, but an oral dose of 400 mg/d given cyclically or continuously has been successfully used in several large studies (8, 9, 12). We therefore used an intravenous dose that was similar to the bioavailable dose of 400 mg/d orally used in previous studies (23, 24), assuming an intestinal absorption rate of 1–2%.
Various intermittent intravenous clodronate regimens comparable with ours have previously been proven to be effective in preventing and treating postmenopausal bone loss (9–11), and other bisphosphonates, such as pamidronate (25) and ibandronate (26), have also been shown to be effective in increasing BMD when given intravenously every 3 months for 1 y. However, the overall negative result of our study may have resulted from differences in the treatment regimens, drug potencies, or patient populations. It is possible that we would have obtained a better response in BMD if we had used a higher dose of clodronate. A study in patients with asthma and corticosteroid-induced bone loss showed a dose-dependent response to clodronate treatment, with a significant increase in spinal BMD but only with oral doses 1600 mg/d (7). Although active treatment with clodronate for 12 mo generally results in a significant increase in BMD (6, 7, 10–13), a few studies reported beneficial effects on BMD that were significant only after 18–24 mo of clodronate therapy (9, 27). Given the results of the current study, it would have been interesting to see what the effect on BMD would have been in our patients if the active treatment had been continued beyond 12 mo.
The exact mechanism underlying bone disease in HPN patients is poorly understood, and the histologic findings in bone biopsy samples vary, although a mineralization defect has been described (1, 28). Any preexistent defect in the mineralization of bone could have accounted for the lack of a significant response to clodronate observed in this study as a result of the inhibitory effect of bisphosphonates on crystal formation in bone. A concern with the use of all bisphosphonates is the potential impairment of bone mineralization, which can result in osteomalacia. However, this is a rare event, which has only been reported with continuous high-dose treatment with etidronate (29).
The effect of bisphosphonates is influenced by the rate of bone turnover, and patients with a high rate of bone turnover respond better to treatment (30). In HPN patients, several histomorphometric studies have shown a low rate of bone turnover and a low rate of bone formation (31, 32); these results are further supported by the results of a recent study of bone-turnover markers (33). However, other studies of bone-turnover markers have not confirmed a low rate of bone formation (34, 35). The patients in our study had a higher rate of bone formation and higher concentrations of bone-resorption markers at baseline, indicating a high rate of bone turnover. Yet, we found no positive correlation between the bone-turnover rate and increases in BMD. Patients characterized by a low or normal pretreatment bone-turnover rate, as judged by biochemical markers of bone metabolism, appeared to have a slightly better response to treatment than did those with a high rate of bone turnover.
An important finding of this study is that treatment with clodronate induced a significant change in indexes of bone turnover. In agreement with the results of previous studies, there was a sharp decrease in markers of bone resorption, which was evident 3 mo after each clodronate infusion. The long-lasting suppressive effect of bisphosphonates on bone turnover is well known and is most likely explained by the long half-life of bisphosphonates in bone, which is estimated to be several years (36). The magnitude of the reduction in the markers of bone resorption observed in the current study is comparable with the results of large trials of bisphosphonate in which a significant effect on bone mass and fracture rate was observed (37, 38). From this point of view, we used a dosage regimen of clodronate that cannot be considered inadequate. The marked increase in osteocalcin, the marker of bone formation, after the initiation of clodronate therapy is unexplained, and a similar response has not been reported previously. As a result of the physiologic process of coupling between resorption and formation at the basic multicellular unit, a decrease in the markers of both remodeling processes is usually seen in bisphosphonate trials (13, 37, 38). The transient change in osteocalcin observed in the current study might have been a chance finding, or it could theoretically be explained by the fact that bisphosphonates may increase the number of osteoblasts and osteocalcin formation, although this has only been documented in osteoblast cultures in vitro (39).
Currently, measures to prevent and treat osteoporosis in HPN patients are being debated. Recently, Buchman and Moukarzel (40) discussed the various therapeutic options for osteoporosis, which included vitamin D supplementation, treatment with calcitonin or parathyroid hormone, and hormone replacement therapy. However, there are not yet official recommendations supported by data from controlled studies in patients with HPN-associated bone disease.
In conclusion, this study showed that cyclical intravenous therapy with 1500 mg clodronate every 3 mo for 1 y is well tolerated and produces significant suppression of bone resorption. We found a nonsignificant trend toward improvement in the BMD of the spine and various skeletal sites, but found no overall significant effect. Nonetheless, the results are encouraging and suggest that antiresorptive therapy may be important in the management of osteoporosis, which is a serious complication of long-term parenteral nutrition. Considering that the population of patients requiring HPN therapy is heterogeneous and that there are a limited number of patients at any one center at a given time, a controlled, prospective, multicenter study of newer bisphosphonates with higher antiresorptive potency is needed.
ACKNOWLEDGMENTS
The technical assistance of Jette Christiansen, Dorte Christensen, Bodil Pedersen, Solveig Petersen, and Jenni Teilmann was greatly appreciated, and we are indebted to the staff at the outpatient clinic for their excellent help with the study.
REFERENCES