Calcified Tissue International

, Volume 88, Issue 5, pp 402–408

No Effect of Rosuvastatin in the Zoledronate-Induced Acute-Phase Response

  • Polyzois Makras
  • Athanasios D. Anastasilakis
  • Stergios A. Polyzos
  • Ilias Bisbinas
  • Grigorios T. Sakellariou
  • Socrates E. Papapoulos
Original Research

DOI: 10.1007/s00223-011-9468-2

Cite this article as:
Makras, P., Anastasilakis, A.D., Polyzos, S.A. et al. Calcif Tissue Int (2011) 88: 402. doi:10.1007/s00223-011-9468-2

The acute-phase response (APR) is frequently observed in patients treated with intravenous (iv) zoledronate (ZOL). We investigated whether a short course of rosuvastatin (ROSU) could attenuate the ZOL-induced APR through blocking the mevalonate pathway at a proximal level. Twenty-eight osteoporotic postmenopausal women with no prior bisphosphonate use (mean age 65.3 ± 1.9 years) were subjected to ZOL iv infusion. Patients were randomly assigned into either a ROSU+ group (n = 12), which received ROSU 10 mg/day starting 5 days before the infusion of ZOL for a total period of 11 days, or a ROSU− group (n = 16), which did not receive ROSU. The visual analog pain scale (VAS) for musculoskeletal symptoms and body temperature was used to define clinically APR. In addition, white blood cell (WBC) count, leukocytic subpopulations, and C-reactive protein (CRP) were obtained before and 48 h following the infusion. Seven (58.3%) patients in the ROSU+ group and 13 (81.3%) in the ROSU− group experienced APR (P = not significant). No difference was found in fever and VAS measurements. CRP and granulocytes increased significantly in both groups; WBC count increased, while lymphocytes and eosinophils decreased significantly only in the ROSU− group. In a post hoc analysis of only patients with an APR, all laboratory parameters exhibited a similar significant change solely within the ROSU− group. In conclusion, our data suggest that a short course of ROS at this dose cannot prevent the ZOL-induced APR among osteoporotic women. Milder changes in acute-phase laboratory parameters in ROSU+ patients suggest that studies with higher doses may be warranted.


Acute-phase response Zoledronate Osteoporosis Statin Rosuvastatin 

Bisphosphonates are the mainstay of treatment of common skeletal disorders such as osteoporosis, metastatic bone disease, and Paget’s disease of bone. Treatment with nitrogen-containing bisphosphonates, especially when given intravenously (iv), may be associated with increases in body temperature and flu-like symptoms such as malaise, myalgia, and bone pains and less frequently ophthalmic reactions such as conjunctivitis and uveitis [1]. These symptoms may be severe, requiring treatment with analgesics or in some cases NSAIDs or prednisone. This response, referred to as the “acute-phase response” (APR), which occurs within 48 h after administration of the bisphosphonate, mostly after the first treatment, is transient, reverses without any specific treatment even upon continuation of the bisphosphonate, and is accompanied by transient decreases in lymphocyte count and increases in serum C-reactive protein (CRP), IL-6, and TNF-α [2, 3, 4]. These biochemical changes are characteristic of an APR but of a lesser magnitude than those induced by an infectious agent [3, 5].

It is currently believed that the APR following the administration of nitrogen-containing bisphosphonates is related to their molecular mechanism of action, namely, the inhibition of farnesyl pyrophosphate synthase (FPPS), an enzyme of the mevalonate pathway which is the target of this class of bisphosphonates [6, 7, 8]. This effect leads to accumulation in blood monocytes of metabolites upstream of FPPS, such as isopentenyl/dimethylallyl diphosphate, which activate γδ T cells and stimulate their proliferation, resulting in increased production of proinflammatory cytokines [9]. In vitro studies reported that inhibition of HMG-CoA reductase, an enzyme higher up in the mevalonate pathway, by statins blocks the synthesis of isopentenyl/dimethylallyl diphosphate and may abolish the activation of γδ T cells [10]. Whether such an intervention may also prevent the development of bisphosphonate-induced APR in vivo is currently unknown. In a recent study of 12 children treated mainly with iv zoledronate (ZOL), atorvastatin given for 3 days failed to prevent an APR as assessed by a visual analog pain scale (VAS) and need for oxycodone and acetaminophen treatment [11]. However, in contrast to previous studies, there were no significant changes in either serum CRP or γδ T cells with or without atorvastatin treatment and changes in body temperature were not specifically reported. Furthermore, it is not known whether these findings in children may also be applicable to elderly persons, the target population of bisphosphonate treatment.

In this study we, therefore, tested the hypothesis that rosuvastatin (ROSU) may alter the frequency or severity of the APR following an iv infusion of ZOL in women with postmenopausal osteoporosis.

Patients and Methods


This was a prospective, open-label study of the efficacy of ROSU treatment to prevent the ARP to a single infusion of ZOL. ROSU is a new, synthetic, potent, hydrophilic statin, as opposed to lipophilic statins used in similar studies (e.g., atorvastatin, fluvastatin). Thirty-two women with postmenopausal osteoporosis defined by a bone mineral density (BMD) T score of ≤–2.5 at the lumbar spine and/or the femoral neck or by the presence of at least one vertebral fracture were recruited from the outpatient clinics of the two participating centers and included in the study. Exclusion criteria were (1) previous treatment with any bisphosphonate, (2) active liver disease, (3) increased levels of serum transaminases or creatinine phosphokinase, (4) creatinine clearance < 60 ml min−1 1.73 m−2, (5) known sensitivity to statins, (6) treatment with any statin during 6 months prior to the study, and (7) fever and/or any viral or bacterial infection 30 days before study.

Patients were randomized into two groups. The first group received ROSU 10 mg/day (ROSU+) starting 5 days before the infusion of ZOL for a total period of 11 days, while the second group did not receive any statin (ROSU−) and served as control. All women received a single iv infusion of ZOL 5 mg between 0800 and 0900 and were supplemented with calcium 1,000 mg/day and vitamin D 800 IU/day starting at least 5 days before the infusion. Steroids, pain medications, or NSAIDs were used only in case of APR and were recorded accordingly. Informed consent was obtained from all patients; the study was approved by the local ethics committees and was in accordance with the Declaration of Helsinki and the International Conference on Harmonization for Good Clinical Practice.


All patients recorded axillary temperature every morning (0800–0900) and every evening (2200–2300) for 5 days prior to the infusion and every 6 h for 3 days following the infusion of ZOL. Patients completed also a VAS at baseline and at 24 and 48 h following the infusion. We defined APR as a rise in axillary temperature above 38°C at any time over the 48 h following the infusion of ZOL and an increase in musculoskeletal pain by more than one grade in the VAS.

Morning fasting blood samples were obtained from all patients at baseline and at 48 h after the infusion for the measurement of white blood cell (WBC) count and CRP, as laboratory indices of APR. Serum samples for CRP determination were stored at −80°C and measured in the same assay at the end of the study. In addition, serum transaminases, calcium, and creatinine clearance were determined at baseline, 24 h, and 48 h.

Statistical Analysis

Data for continuous variables are presented as mean ± standard error of the mean (SE). Data for categorical variables are presented as numbers. The chi-squared test or Fischer’s exact test was used to identify differences between categorical variables. The Mann-Whitney test was used to identify differences between two independent groups of continuous variables. Wilcoxon’s signed ranks test was used to identify differences within the levels of continuous variables, if there were two repeated measurements. Repeated-measures ANOVA was used to identify differences within the levels of continuous variables, if there were more than two measurements. Adjustment for violations of the sphericity assumption was performed with the Greenhouse-Geisser correction. In case of significant differences in repeated-measures ANOVA, Bonferroni post hoc analysis was used for multiple pairwise comparisons. P < 0.05 was considered statistically significant in all of the above tests. Statistical analysis was performed with SPSS 13.0 for Windows (SPSS, Inc., Chicago, IL). Power analysis was performed with GPower 3.0.8 (University of Kiel, Kiel, Germany).


Of the 32 patients recruited in the study, four (all ROSU+) withdrew their consent before the start of the treatment protocol. Thus, 12 patients in the ROSU+ group and 16 patients in the ROSU− group participated and completed the study. Age, BMD T scores, and body mass index (BMI) did not differ between the two groups, being 65.2 ± 1.7 and 65.4 ± 2.6 years, −2.66 ± 0.08 and −2.92 ± 0.18, and 28.9 ± 1.3 and 28.2 ± 1.4 kg m−2 in the ROSU+ and ROSU− groups, respectively.

ZOL treatment induced an APR in 20/28 (71.4%) studied patients (14 within 24 h and six in 48 h). There were 7/12 (58.3%) patients in the ROSU+ group and 13/16 (81.3%) patients in the ROSU− group; the difference between the two groups was not significant (NS). Mean temperature at 24 h was 38.1 ± 0.4°C in the ROSU+ group and 38.0 ± 0.3°C in the ROSU− group (P = NS) and 37.3 ± 0.3°C and 37.5 ± 0.3°C at 48 h, respectively (P = NS). Similarly, pain increased significantly (P = 0.001 and P < 0.001, respectively) with treatment, but there were no differences between the two groups at either 24 or 48 h (ROSU+ baseline 1.6 ± 0.4, 24 h 4.7 ± 1.0, 48 h 3.1 ± 0.8; ROSU− baseline 0.7 ± 0.3, 24 h 5.4 ± 0.7, 48 h 4.4 ± 0.8). All patients with an increase in pain also had fever. No patient reported ophthalmic or gastrointestinal symptoms. Rescue medications (steroids, pain medications, NSAIDs) were equally used between ROSU+ and ROSU− patients with an APR (data not shown).

Changes in laboratory parameters were consistent with the induction of an APR (Table 1). Serum CRP increased significantly in both groups (Fig. 1), and there were no significant differences between the two groups at 48 h. WBC count increased in both groups, but the increase was significant only in the ROSU− group. Granulocytes increased significantly in both groups, without any difference between the two groups at 48 h, and lymphocytes decreased (Fig. 2); but this decrease was significant only in the ROSU− group, being also different from ROSU+ at 48 h.
Table 1

Laboratory parameters at baseline and 48 h






48 h


48 h

Patients (N)





CRP (mg l−1)

3.1 ± 0.8

55.5 ± 23.6*

4.7 ± 2.4

64.6 ± 36.6*

WBC (μl−1)

6,840 ± 596

7,150 ± 770

5,600 ± 406

6,989 ± 646*

Granulocytes (μl−1)

3,960 ± 477

4,840 ± 746*

3,083 ± 262

5,487 ± 695*

Lymphocytes (μl−1)

2,256 ± 177

1,751 ± 242

1,965 ± 190

1,031 ± 132*,**

Eosinophils (μl−1)

135 ± 19

81 ± 19

123 ± 24

55 ± 18*

Granulocytes (%)

57 ± 3

66 ± 5

55 ± 2

76 ± 3*

Lymphocytes (%)

34 ± 3

26 ± 5

35 ± 2

17 ± 3*

Eosinophils (%)

1.9 ± 0.2

1.1 ± 0.2

2.3 ± 0.5

0.9 ± 0.3*

ROSU rosuvastatin; N number; WBC white blood count

P < 0.05 compared to baseline, ** P < 0.05 compared to ROSU+

Fig. 1

Individual alterations of CRP levels in both groups. White squares indicate patients with an acute-phase response (APR), while black squares correspond to patients without an APR

Fig. 2

Individual alterations of lymphocytes in both groups. White squares indicate patients with an acute-phase response (APR), while black squares correspond to patients without an APR

These results suggested that ROSU treatment, though not effective at preventing the APR to ZOL, may affect to some extent the hematological changes associated with the APR. To investigate this further, we performed a post hoc exploratory analysis of the results only of patients who experienced an APR. As shown in Table 2, the mean serum CRP value of ROSU− at 48 h was nearly twofold higher than that of ROSU+, as was the percentage decrease in lymphocytes (53 vs. 30%). In addition, granulocytes increased by 93% in ROSU− versus 11.1% in ROSU+.
Table 2

Characteristics and laboratory parameters of patients with an acute-phase response following intravenous zoledronate






48 h


48 h

Patients (N)





Age (years)

63.9 ± 2.6

64.9 ± 2.8

BMI (kg m−2)

30.5 ± 1.5

27.2 ± 1.6

BMD L2–L4 (g cm−3)

−2.70 ± 0.12

−2.95 ± 0.21

CRP (mg l−1)

3.6 ± 1.4

46.8 ± 34.1

5.3 ± 3.0

82.5 ± 46.6*

WBC (μl−1)

7,940 ± 833

7,675 ± 983

5,755 ± 456**

7,496 ± 725*

Granulocytes (μl−1)

5,007 ± 629

5,563 ± 823

3,126 ± 264**

6,064 ± 802*

Lymphocytes (μl−1)

2,216 ± 286

1,550 ± 325

2,079 ± 216

967 ± 159*

Eosinophils (μl−1)

146 ± 25

98 ± 22

118 ± 29

33 ± 12*,**

Granulocytes (%)

63 ± 3

73 ± 4

55 ± 2**

79 ± 4*

Lymphocytes (%)

28 ± 3

20 ± 3

36 ± 2

15 ± 3*

Eosinophils (%)

1.8 ± 0.2

1.3 ± 0.2

2.1 ± 0.6

0.5 ± 0.2*,**

ROSU rosuvastatin; BMD bone mineral density; BMI body mass index; N number; WBC white blood count

* P < 0.05 compared to baseline, ** P < 0.05 compared to ROSU+ (between-group comparison, Mann–Whitney test)

AT 48 h there were no differences in serum transaminases and calcium concentrations or in creatinine clearance between the two groups and all values were within their respective reference ranges.


We show here that short-term treatment with ROSU of women with postmenopausal osteoporosis, naive to bisphosphonate treatment, does not prevent the APR induced by a single iv infusion of the nitrogen-containing bisphosphonate ZOL.

The occurrence of an APR to treatment with nitrogen-containing bisphosphonates has been known for about 30 years [12], but its true incidence is unknown due mainly to different definitions used to describe it and the regular use of acetaminophen in some studies [11, 13, 14, 15]. In our study, using conservative criteria to determine the APR, 71.4% of the patients treated with ZOL experienced an APR. None of them had ever been treated with a nitrogen-containing bisphosphonate, and they did not receive any symptomatic treatment before the development of symptoms. A similar incidence (63%) has, however, been previously reported in patients with Paget’s disease treated with iv pamidronate [5] using only the rise in body temperature to define the APR. In our study, no patient without fever experienced significant musculoskeletal pain. The response was further associated with the changes previously described during a bisphosphonate-induced APR such as increases in serum CRP and granulocytes and decreased lymphocytes [2, 4, 5]. Most studies with bisphosphonates lack this information, and the identification of an APR in clinical trials depends mainly on clinical parameters, which may be related to the response and are collected as adverse events.

In clinical practice, patients are symptomatically treated after the development of an APR, but there are also suggestions to treat all patients receiving for the first time iv nitrogen-containing bisphosphonates with paracetamol/acetaminophen [1]. The experimental evidence linking the molecular mechanism of action of nitrogen-containing bisphosphonates to the induction of the APR led to the suggestion that statins which inhibit the mevalonate pathway upstream of FPPS may be useful in preventing the APR by inhibiting the synthesis of isopentenyl/dimethylallyl diphosphate [16]. In addition, some statins have been reported to increase serum 25-hydroxyvitamin D concentrations within 8 weeks of start of treatment [17, 18], and they are also known to decrease serum CRP [19], providing, thus, additional arguments for a potential attenuation of the APR in bisphosphonate-treated patients. In our study ROSU starting 5 days before the infusion and given for a total of 11 days did not prevent the clinical expression of the APR to ZOL, a finding which is in agreement with a study of children given atorvastatin for 3 days [11] and a recent study with postmenopausal patients given fluvastatin 40 mg either once or three times [20]. These results suggest that the type of statin (hydrophilic ROSU or lipophilic atorvastatin, fluvastatin) is not responsible for the lack of an effect. Similarly, we could not detect any differences in the biochemical changes of the APR between patients who were treated with ROSU and those who were not. Thus, despite the number of potential mechanisms by which statins may affect the development of a bisphosphonate-induced APR, human studies do not support a role of statins in the prevention of the response. The reason for that may lie in the clear differences between in vitro and in vivo studies. In in vitro studies statins are added directly in the culture, whereas in vivo they are rapidly metabolized in the liver and do not have sufficient time to target monocytes or bone cells. This also is the reason for the lack of an effect of statins on bone metabolism in vivo, despite their clear efficacy at reducing osteoclastic resorption in vitro [21]. However, in a post hoc analysis of only patients who experienced a clinical APR, we found a clear trend for milder changes in CRP and blood counts in the patients who received ROSU, which, however, was not sufficient to prevent it. A similar trend was observed only for IL-6 levels in the study of fluvastatin but not for CRP and other proinflammatory cytokines [20]. A higher dose of ROSU might have been effective, but the question is whether such an intervention can be beneficial for the patient due to the well-known dose-dependent side effects of statins. At the individual level, marked changes in leukocytic subpopulations or CRP were not always associated with the development of an APR (Figs. 1, 2). This was also the case in the fluvastatin study, where great increases of proinflammatory cytokines did not always trigger an APR, thereby indicating a more complex relationship between cytokines and APR symptoms [20]. The length of treatment with a statin, as applied in our study, does not seem to play a role. In a recently reported post hoc analysis of the HORIZON trial in which postmenopausal women with osteoporosis were treated with yearly iv infusions of ZOL, chronic statin use did not affect the development of an APR assessed by analysis of relevant symptoms [1].

The finding that not all patients treated for the first time with a nitrogen-containing bisphosphonate develop an APR and the “memory” of the nitrogen in the bisphosphonate structure in repeated treatments has not yet been resolved. It has been clearly established that the induction of an APR in patients previously treated with the same or another nitrogen-containing bisphosphonate is less common despite the continuous effect of the bisphosphonate on FPPS [1]. Thus, despite the progress in our understanding of the mechanism and the determinants of the bisphosphonate-induced APR, there remain unanswered questions. For the practicing clinician awareness of this adverse event is essential as the clinical response can sometimes be very severe, requiring treatment even with glucocorticoids. On the other hand, the reaction is transient and commonly does not occur upon retreatment, which is reassuring.

An intriguing, recently reported finding is the association between a low serum 25-hydroxyvitamin D concentration and the development of a bisphosphonate-induced APR [13]. In addition, there were significant negative relationships between serum 25-OHD and body temperature or CRP. These observations are in line with earlier studies showing higher basal plasma concentrations of PTH in patients with Paget’s disease with a febrile response to pamidronate treatment compared to those without such a response and a highly significant correlation between basal PTH concentrations and peak IL-6 levels after treatment [3]. The mechanism(s) underlying these associations is currently unknown. We did not measure serum 25-OHD concentrations in our study, but all patients received vitamin D and calcium for at least 5 days before the infusion of ZOL.

In conclusion, our results show that ROSU, as used in this study, does not prevent the ZOL-induced APR in osteoporotic women. The smaller changes in acute-phase laboratory parameters in patients treated with ROSU suggest that studies with higher doses may be warranted. However, such studies will only confirm or refute the mechanistic hypothesis rather than offer a clinically useful intervention.

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • Polyzois Makras
    • 1
  • Athanasios D. Anastasilakis
    • 2
  • Stergios A. Polyzos
    • 3
  • Ilias Bisbinas
    • 4
  • Grigorios T. Sakellariou
    • 5
  • Socrates E. Papapoulos
    • 6
  1. 1.Department of Endocrinology and Diabetes251 Hellenic Air Force and VA General HospitalAthensGreece
  2. 2.Department of Endocrinology424 Military HospitalThessalonikiGreece
  3. 3.Second Department of Internal MedicineAristotle University of Thessaloniki, Hippokration General HospitalThessalonikiGreece
  4. 4.First Department of Orthopaedics424 Military HospitalThessalonikiGreece
  5. 5.Department of Rheumatology424 Military HospitalThessalonikiGreece
  6. 6.Department of Endocrinology and Metabolic DiseasesLeiden University Medical CenterLeidenThe Netherlands

Personalised recommendations