Continuous Parathyroid Hormone Induces Cortical Porosity in the Rat: Effects on Bone Turnover and Mechanical Properties


  • The authors have no conflict of interest


We examined the time course effects of continuous PTH on cortical bone and mechanical properties. PTH increased cortical bone turnover and induced intracortical porosity with no deleterious effect on bone strength. Withdrawal of PTH increased maximum torque to failure and stiffness with no change in energy absorbed.

Introduction: The skeletal response of cortical bone to parathyroid hormone (PTH) is complex and species dependent. Intermittent administration of PTH to rats increases periosteal and endocortical bone formation but has no known effects on intracortical bone turnover. The effects of continuous PTH on cortical bone are not clearly established.

Materials and Methods: Eighty-four 6-month-old female Sprague-Dawley rats were divided into three control, six PTH, and two PTH withdrawal (WD) groups. They were subcutaneously implanted with osmotic pumps loaded with vehicle or 40 μg/kg BW/day human PTH(1-34) for 1, 3, 5, 7, 14, and 28 days. After 7 days, PTH was withdrawn from two groups of animals for 7 (7d-PTH/7d-WD) and 21 days (7d-PTH/21d-WD). Histomorphometry was performed on periosteal and endocortical surfaces of the tibial diaphysis in all groups. μCT of tibias and mechanical testing by torsion of femora were performed on 28d-PTH and 7d-PTH/21d-WD animals.

Results and Conclusions: Continuous PTH increased periosteal and endocortical bone formation, endocortical osteoclast perimeter, and cortical porosity in a time-dependent manner, but did not change the mechanical properties of the femur, possibly because of addition of new bone onto periosteal and endocortical surfaces. Additionally, withdrawal of PTH restored normal cortical porosity and increased maximum torque to failure and stiffness. We conclude that continuous administration of PTH increased cortical porosity in rats without having a detrimental effect on bone mechanical properties.


HYPERPARATHYROIDISM IS ASSOCIATED with decreased BMD, particularly at sites containing predominantly cortical bone.(1–3) In contrast, the density of cancellous bone in vertebrae is well preserved.(3) Thus, bone densitometry studies suggest that the response to elevated parathyroid hormone (PTH) levels is site specific. Similar site specificity was observed in postmenopausal osteoporosis patients who received intermittent administration of PTH.(4–6)

Cortical bone contributes significantly to the mechanical strength of an individual bone.(7–9) Small changes in microstructure of compact bone exert a more pronounced effect on bone strength than similar changes in cancellous bone would create.(10,11) A loss of cortical bone can involve either thinning of the cortex or an increase in cortical porosity. The cortical bone response to intermittent PTH has been investigated in several animal models including dogs,(12) ferrets,(13) rabbits,(14) monkeys,(15) and rats.(16) The impact of cortical porosity on the mechanical properties of cortical bone after intermittent PTH treatment remains controversial. Intermittent PTH increases cortical porosity of tibia and humerus in rabbits and monkeys, respectively, by stimulating intracortical remodeling. The increased cortical porosity in rabbits was accompanied by periosteal and endocortical bone formation so that mechanical strength was increased. In monkeys, PTH stimulates intracortical remodeling, which results in increased porosity, mostly localized toward the endocortical surface, producing no deleterious effect on biomechanical properties. After PTH was discontinued in monkeys, the increased cortical porosity was resolved in the lower-dose group but remained in the higher-dose group. Increased cortical porosity has not been reported in rats treated with intermittent PTH.

Intermittent administration of PTH increases cortical bone mass in rats, whereas continuous infusion of the same dose induced hypercalcemia and death.(17,18) Normal rats lack Haversian canals to remodel cortical bone and exhibit cumulative appositional cortical bone growth when PTH is given at dose rates that do not cause hypercalcemia. However, there is evidence that continuous PTH induces cortical bone loss in intact(19) and ovariectomized rats.(20) Thus, it may be possible to develop a rat model to investigate the effects of PTH treatment and withdrawal on cortical bone turnover and biomechanical properties.

This experiment was designed to evaluate the time course effects of continuous PTH on cortical bone turnover, intracortical porosity, and bone strength and to study the consequences of withdrawing PTH. Specific questions included the following: (1) does continuous PTH alter bone formation on the periosteal and/or endocortical surfaces; (2) does PTH increase cortical porosity, and if so, is the distribution of porosity uniform throughout the diaphysis of a long bone; (3) what is the impact of continuous PTH treatment on bone mechanical properties; and (4) what effect does withdrawal of PTH have on bone turnover and mechanical properties.


Preliminary experiment

We investigated the effects of intermittent treatment with human PTH 1-34 (Bachem, Torrance, CA, USA) at the dose of 80 μg/kg BW/day on cortical bone histomorphometry of tibias in ovariectomized adult Sprague-Dawley rats (Harlan Sprague-Dawley, Indianapolis, IN, USA). The rats were sham operated or ovariectomized at 3 months of age and remained untreated for 4 weeks. They were then injected subcutaneously with PTH 6 days/week for 5, 10, and 15 weeks. In addition, we performed μCT analysis on rat tibias after 15 weeks of PTH treatment. The studies have been described in detail elsewhere.(21)

Main experiments

Rats were housed three per cage and maintained under 12:12 h light/dark cycle with water and standard rat chow (Purina laboratory rodent diet 5001; Ralston-Purina, St Louis, MO, USA) ad libitum. All procedures performed throughout the experiment conformed to the guidelines of the Institutional Animal Care and Use Committee at the Mayo Foundation.

Time course effects of continuous PTH

This experiment was performed to establish the time course effects of continuous PTH treatment on cortical bone turnover and porosity and to determine the consequences of withdrawing PTH on mechanical properties. Eighty-four adult female intact rats ∼6 months of age were divided into three experimental studies: 7-, 14-, and 28-day study (Table 1).

Table Table 1.. Experimental Design
original image

Seven-day study

Rats were subdivided into five groups, providing one control and four PTH groups. They received subcutaneously implanted 7-day osmotic pumps (DURECT, Cupertino, CA, USA) containing either vehicle or 40 μg/kg/day PTH. A control group was killed at the end of experiment, whereas the PTH groups were killed after 1, 3, 5, and 7 days of treatment. Fluorochromes were injected juxta-tail vein. Tetracycline (20 mg/kg) was given on the first day of the experiment in both the control and 7d-PTH groups and calcein (20 mg/kg) was given 1 day before death in every group to label mineralizing bone.

Fourteen- and 28-day studies

Rats were subdivided into three groups for each study: control, PTH, and PTH followed by PTH WD (Table 1). They received subcutaneously implanted 14-day osmotic pumps containing either vehicle or PTH at the same dose as in the 7-day study. A new pump was replaced after 2 weeks in the 28-day study. Calcein and tetracycline were given on days 8 and 1 before death, respectively.

At the end of experiment, animals were anesthetized with ketamine HCl (50 mg/kg BW):xylazine HCl (5 mg/kg BW) and killed by cardiectomy. Right tibias were quickly excised and fixed in 70% ethanol for bone histomorphometry and μCT analysis. Right femora from the 28-day study were removed and kept at −70°C for biomechanical testing.

Bone preparation and histomorphometry

Cross-sections (150 μm thick) were cut at a site proximal to tibiofibular junction with a low speed Isomet saw (Buehler, Lake Bluff, IL, USA). Two sections from each animal were ground to an approximate thickness of 20 μm and mounted unstained for microscopic examination of cortical bone. Histomorphometric procedures were carried out using a semiautomatic image analysis system (Osteometric, Atlanta, GA, USA). All parameters followed the standard nomenclature proposed by Parfitt et al.(22) Measurements included cross-sectional area (CSA), defined as the total area surrounded by periosteal surface; medullary area (MA), defined as an area of the medullary space; cortical area (CA), calculated as the difference between cross-sectional and medullary area; and single-labeled perimeter (sL.Pm), defined as the length of the last label at endocortical and periosteal surface. Bone formation rate (BFR) was calculated as the bone area between the calcein and tetracycline labels divided by the labeling period. Pore area was measured to calculate cortical porosity as a percentage of cortical bone area as described.(15,20)

The diaphysis was cut at 1 cm proximal to the tibiofibular synostosis and fixed in 70% ethanol for μCT scans.

The remaining proximal tibias were dehydrated in series of increasing concentrations of ethanol, infiltrated, embedded without demineralization in glycol methylmethacrylate, and sectioned longitudinally at 5 μm on a Reichert-Jung Supercut 2050 microtome. The sections were stained with acid phosphatase to identify osteoclasts. Osteoclast surface along endocortical bone at 1 mm distal to the growth plate was determined and expressed as a percentage of endocortical bone surface.


Tibias were embedded in paraffin and analyzed on a μCT scanner, as described previously in detail.(23) The images were reconstituted with a (20-μm)3 voxel size.

Biomechanical testing

Femora were thawed at room temperature before examination of the biomechanical properties. Proximal and distal ends were placed in aluminum fixtures, leaving the diaphysis free for torsion testing. They were tested to failure on a custom biaxial electromechanical testing machine (Mayo Clinic). Maximum torque (N-mm) was quantified as the maximum force encountered before failure. Stiffness (N-mm/degree), which reflects the sample's resistance to rotation, was calculated as the slope of the tangent line of the force-rotation curve. Ultimate strength (N/mm2) was calculated according to the guidelines.(24)

Statistical analysis

Data are expressed as mean ± SE. Multiple comparisons were analyzed by one-way ANOVA followed by Fisher's protected least significant difference posthoc test on determination of significance. Significance was established at p < 0.05 comparing all treatment groups to control, 7d-PTH/7d-WD to 14d-PTH, and 7d-PTH/21d-WD to 28d-PTH.


Preliminary experiment

Increased cortical bone porosity was not detected by histomorphometry in intact or ovariectomized rats treated with intermittent PTH. Similarly, μCT analysis failed to detect increases in cortical porosity in ovariectomized rats treated with intermittent PTH for 15 weeks (data not shown).

Table Table 2.. Time Course Study of PTH on Body Weight and Static Cortical Bone Histomorphometric Measurements
original image

Main experiments

Data from 7-, 14-, and 28-day control groups did not differ from one another and were pooled. The time course effects of PTH on body weight and static cortical bone histomorphometry are shown in Table 2. All rats lost weight (21%) during the first week of PTH treatment. Body weight rebounded and was restored to the control value within 14 days. PTH treatment and PTH withdrawal, regardless of duration, did not affect tibial cross-sectional and cortical area. Twenty-one days after withdrawal of PTH, marrow area was significantly decreased relative to control rats.

A 7-day interlabel interval did not provide adequate separation to accurately measure double-labeled surface at the periosteum. Single-labeled (last label) perimeter at 1 day before death was used as an index for changes in bone formation at the periosteum (Fig. 1). 28d-PTH increased periosteal single-labeled perimeter. After 7d-PTH treatment followed by 7d-WD, the single-labeled perimeter decreased compared with control. However, the labeled perimeter was equivalent to control value after 7d-PTH/21d-WD. PTH treatment increased single-labeled perimeter at the endocortical bone surface within 3 days (Fig. 2). In contrast to periosteum, endocortical single-labeled perimeter in PTH WD rats, irrespective of time point, was higher than control rats. The single-labeled perimeter remained increased after 7d-WD. However, the value was equivalent to 7d-PTH value after withdrawal of PTH for 21 days.

Figure FIG. 1..

Periosteal single-labeled perimeter (mm) in the tibial diaphysis of control, PTH, and PTH followed by PTH withdrawal. Each bar is mean ± SE. ap < 0.05 compared with control;bp < 0.05 compared with 28d-PTH.

Figure FIG. 2..

Endocortical single-labeled perimeter (mm) in the tibial diaphysis of control, PTH, and PTH followed by PTH withdrawal. Each bar is mean ± SE.ap < 0.05 compared with control;bp < 0.05 compared with 28d-PTH.

Figure 3 shows BFR at the endocortical bone surface. In contrast to periosteum, a 7-day interlabel interval was adequate to measure double label on the endocortical surface in some PTH-treated groups. The BFR was, however, below detection in the control and 7d-PTH groups. The endocortical BFR was increased in PTH-treated rats because of increases in both double-labeled surface and mineral apposition rate (data not shown), and the levels decreased to near control levels after withdrawal of PTH.

Figure FIG. 3..

Endocortical bone formation rate (mm2 × 10−3/day) in the tibial diaphysis of control, PTH, and PTH followed by PTH withdrawal. Each bar is mean ± SE.ap < 0.05 compared with control;bp < 0.05 compared with 28d-PTH.

The effects of PTH and PTH withdrawal on intracortical porosity are plotted as a function of time in Fig. 4. PTH treatment induced a time-dependent increase in cortical porosity. After PTH was withdrawn, the increased porosity disappeared. Figure 5 shows the appearance of the cortical porosity when viewed by fluorescence microscopy and shows the presence of bone formation surrounding pore surfaces, providing evidence for cortical bone remodeling. PTH induced intracortical remodeling in rats, and the extent of remodeling increased with the duration of PTH treatment.

Figure FIG. 4..

Time course effect of PTH (•) and PTH withdrawal (○) on intracortical porosity (% cortical area). Data expressed as mean ± SE. *p < 0.05 compared with control (0d); #p < 0.05 compared with 28d-PTH.

Figure FIG. 5..

Unstained sections of cortical bone under fluorescence microscope from (A) control and (B) 28d-PTH showing morphology of cortical porosity and fluorochrome labeling.

μCT analysis of multiple sites revealed that the PTH-induced cortical porosity was primarily localized close to the endocortical surface (Fig. 6), but showed pronounced regional differences in distribution. Furthermore, the pores formed a continuous channel to the endocortical bone surface. However, channels were found to be continuous with the periosteal surface at the proximal diaphysis. Most pores were located at the medial side of the cross-section in the distal diaphysis, whereas the porosity occurred at the lateral side in the proximal diaphysis. Pores were filled in at all sites in the diaphysis after 21d-WD.

Figure FIG. 6..

μCT images of cortical bone at three different regions of right tibial diaphysis in control, 28d-PTH, and 7d-PTH/21d-WD. Arrows in control group denote a nutrient artery located at dorso-medial side of tibial diaphysis.

Osteoclast perimeter was measured on the endocortical surface, a putative initiation site for PTH-induced cortical porosity. PTH significantly increased osteoclast perimeter per endocortical perimeter (%) within 1 day of treatment (0.58 ± 0.15 versus 10.04 ± 3.91, p < 0.001), and osteoclast perimeter remained elevated throughout the duration of PTH treatment. The maximum value was observed in 28d-PTH (13.76 ± 2.28%), but this value did not differ significantly from 1 day-treated animals. The osteoclast perimeter decreased after withdrawal of PTH for 21 days (4.52 ± 1.28%) but remained elevated compared with the control (p < 0.05).

The effects of PTH and PTH withdrawal on mechanical properties are shown in Table 3. PTH had no effect on any parameter of bone strength. In contrast, 7d-PTH/21d-WD increased maximum torque and stiffness compared with control rats by 77% and 139%, respectively.

Table Table 3.. Effect of Continuous PTH on Bone Strength
original image


PTH is approved for the treatment of established postmenopausal osteoporosis. An ongoing concern of treatment is that increased cortical porosity may lead to reduced cortical bone mass and strength. In this study, we examined how PTH-induced increased cortical bone turnover in rats influences bone mechanical properties by delivering PTH continuously followed by PTH withdrawal. Our results suggest that a continuous elevation of PTH significantly increases cortical porosity, mainly near endocortical surfaces. Because the increased porosity is accompanied by new periosteal and endocortical bone formation, it causes no deleterious effect on bone strength. Withdrawing PTH altered the bone mechanical properties by increasing maximum torque to failure and stiffness. As a consequence, the bone became more brittle.

Continuous PTH treatment increased bone formation at the endocortical surface within 3 days of initiating treatment and at the periosteal surface within 28 days. The response of periosteum to intermittent PTH is also less than that of endocortical bone(21,25,26); PTH increased periosteal mineralizing surface by 2-fold, whereas up to 4-fold increase occurred at the endocortical surface.(21) Although the continued surface accumulation of matrix on periosteal and endocortical surfaces thickened the cortex and increased cortical area with 10 weeks of intermittent PTH treatment in aged ovariectomized rats,(27,28) 28 days of continuous PTH was an insufficient duration of treatment to measure an increase in cross-sectional, cortical, and marrow area in intact rats. A higher single-labeled perimeter in 7d-PTH/7d-WD animals compared with the 7d-PTH group indicated a further increase in bone formation after PTH withdrawal on the endocortical surface. This finding indicates that increased bone formation, once initiated, does not require the continuous presence of elevated PTH.

The effects of PTH on cortical bone porosity have been investigated in several animal models with intracortical Haversian remodeling. Once-daily administration of PTH for 140 days activated intracortical remodeling and increased cortical porosity from 1.3% to 6.3% in rabbit tibial midshaft.(29) Cortical bone turnover is rarely investigated in rats because of their very low baseline levels of osteonal bone remodeling. Cortical porosity represents only 0.1-1.5%(30–32) of cortical bone volume in rats, whereas about 8% and 20% porosity were observed in 40- and 80-year-old humans,(33,34) respectively. We are not aware of reports of increased cortical porosity in rats treated with intermittent PTH. Near-lifetime treatment of intact rats with intermittent PTH stimulated mineral apposition onto both endocortical and periosteal surfaces, resulting in the massive accumulation of bone in femora.(35) We verified by μCT analysis that intermittent PTH does not induce cortical bone remodeling; PTH treatment for 15 weeks increased cortical thickness but did not increase cortical porosity. However, this study clearly shows the induction of cortical bone remodeling and porosity in rats when PTH levels are continuously elevated. It is not clear why rats differ from larger mammals in requiring continuous PTH to stimulate cortical bone remodeling. The differential skeletal response to intermittent and continuous PTH is more likely related to the duration of exposure rather than peak blood levels of the hormone.(17) As a consequence, there is a continuum of side effects as the duration of exposure to PTH is lengthened.

The use of μCT allowed us to visualize in 3D the microarchitecture of bone. Most porosity was localized to endocortical surfaces and more pronounced in the medial and lateral regions of the distal and proximal diaphysis, respectively. However, periosteal porosity was observed at both ventral and dorsal sides near the proximal metaphysis (data not shown). The explanation for the regional variation in porosity distribution along rat long bones remains to be established. The mechanism of increased cortical porosity originates from increased frequency of activation of Haversian systems creating more tunnels. The increase in osteoclast activity then reduces the ability to replace bone removed during remodeling, leading to an increase in Haversian canals or larger tunnels. Several lines of evidence suggest that the cortical porosity induced in rats by continuous PTH treatment is similar to Haversian remodeling. The porosity was continuous with a bone surface. An increase in osteoclast perimeter on the endocortical surface preceded the increase in cortical porosity, suggesting that cortical bone remodeling was initiated on that surface. Erosion pits located on the periosteal surface of the proximal diaphysis suggest that cortical remodeling can be initiated from the periosteum as well. Fluorochrome labels along the perimeter of the intracortical pores indicate that the initial resorptive response is followed by bone formation.

The fragility of bone is dependent on material properties, microstructure, geometry, and mass. The biomechanical strength of whole bone can be evaluated by three- or four-point bending or torsional testing. Because rat long bones are slightly curved, the orientation of bone during bending testing may reduce the precision of the testing.(36) In contrast, torque is constant throughout the tested bone, and geometry of bone has little effect on torsion strength. As observed in monkeys, the increase in cortical porosity with no effect on torsion at failure in long-term PTH treatment implies that changes in porosity alone do not have detrimental effects on mechanical properties. This may be because pores close to the endocortical surface have less effect on mechanical properties than pores near the periosteal surface.(15) Therefore, changes in the spatial distribution of porosity are one of the factors contributing to mechanical function of bone. Also, the addition of new bone onto the endocortical and periosteal surfaces may counteract the detrimental effects of increased porosity. 7d-PTH/21d-WD improved rigidity and stiffness of bone. The restructuring of bone without any effect on energy absorbed indicates that the bone in 7d-PTH/21d-WD animals is more brittle (i.e., has less elasticity and deforms less before fracture). The mechanisms by which withdrawing PTH resulted in the increase in torque and stiffness are not known.

In conclusion, continuous administration of PTH increases cortical porosity primarily close to the endocortical surface. Because the porosity is not uniformly distributed, the impact on the bone mechanical properties is negligible. Withdrawal of PTH, on the other hand, increases rigidity and stiffness of cortical bone, and as a consequence, makes the bone more brittle.


The authors thank Patricia Beighley for embedding the diaphysis, Steven Jorgensen for scanning the diaphysis, and Denise Reyes for reconstructing μCT images and for doing the porosity analysis. We also thank Peggy Backup and Lori Rolbiecki for editorial assistance. This study was supported by NIH Grant AR048833, μCT Grant EB000305, and the Mayo Foundation.