Men's Osteoporosis Support Group

Important study on osteonecrosis of the jaws

J Oral Maxillofac Surg. 2007 Dec;65(12):2397-410, Oral bisphosphonate-induced osteonecrosis: risk factors, prediction of risk using serum CTX testing, prevention, and treatment. Marx RE, Cillo JE J, Ulloa JJ. PMID: 18022461. This is a very important study on the topic of bisphosphonate-induced osteonecrosis of the jaws (BIONJ), a rare but potentially serious complication from taking intravenous (IV) or oral bisphosphonates such as Fosamax (alendronate) or Actonel (risedronate) or ibandronate (Boniva). See a recent Update on this topic for additional information and links to previous topic updates.

Before this study there was considerable doubt as to exactly what factors influenced the onset and resolution of BIONJ, and that was troublesome. It was known that most cases were induced by IV bisphosphonates such as pamidronate (Aredia) and zoledronate (Zometa) used to treat metastatic cancer in the bone. However, a smaller number of cases were definitely linked to oral bisphosphonate medications. Which people were are greater risk, whether to stop the oral bisphosphonate once the BIONJ was detected, and the best way to treat the BIONJ were all unknowns. The Marx and others study has given us answers to many of these questions now, which should provide comfort and guidance for those of us with osteoporosis who are taking a bisphosphonate, either oral or IV.

The abstract linked above covers the details of the study's findings very well, so please read it. I will re-emphasize certain topics mentioned in the abstract and highlight others only mentioned in the complete paper in what follows.

The investigators in this study have data on 116 cases of IV bisphosphonate-induced ONJ with which they could compare and contrast the 30 cases of oral bisphosphonate-induced ONJ. The authors defined BIONJ according to the American Association of Oral and Maxillofacial Surgeons Task Force as, "The presence of nonhealing exposed bone in the maxilla or mandible that has persisted for more than 8 weeks in a patient who has received a systemic bisphosphonate but has not received local radiation therapy."

Results: Twenty-seven cases (90%) resulted from Fosamax use with a mean duration of drug exposure of 5.7 years. Three cases were from Actonel with a mean duration of drug use of 5.0 years. All cases were in women with a mean age of 64.8 years. Sixteen patients (53.3%) were taking the medication for osteopenia, ten (33.%) for osteoporosis and 7 (13.4%) were taking medications due to steroid-induced osteoporosis. Twenty-nine cases (96.7%) were in the mandible and 1 case (3.3%) was in the maxilla. Two cases resulted due to a dental implant procedure. Fifteen cases (50%) resulted spontaneously without a recognized initiating event. Patients had taken a bisphosphonate from 3.3 to 10.2 years total duration with a composite mean of 5.6 years.

Mean size of osteonecrosis and duration of medication use.

Duration

Number

Size

Pain

Infection

3-4 years

8

0.6 cm2

2

2

4-5 years

6

1.5 cm2

3

3

5-6 years

5

3.8 cm2

3

3

6-7 years

3

4.7 cm2

2

3

>7 years

8

7.9 cm2

8

8

Notice that pain was highly correlated with infection: 19/30 (63.3%) patients reported pain associated with the bone exposure. Pain was then correlated with clinical evidence of infection, such as, erythema, swelling, drainage, and/or osteolysis in 18 of 19 patients (94.7%) who reported pain. Also notice that the lesion size was highly correlated with the duration of medication use.

Fasting serum CTX bone turnover marker. See this Update for more on using biochemical markers. The authors' use of the fasting serum CTX marker is one of the most exciting and useful aspects of this study. Their results show that using this marker will provide information as to how at risk an individual might be for BIONJ, and, if you get BIONJ, how to use CTX to determine the degree of bone turnover occurring which will help determine when to treat and what type of treatment might be needed. In summary, this marker tells how much bone is being resorbed and replaced. So a high number indicates greater bone turnover and a low number means little bone is turning over. People at risk for BIONJ would not want to have a very low number.

Seventeen patients were taking a bisphosphonate when first seen. The fasting serum CTX for these people ranged from 30 pg/ml to 102 pg/ml with a mean of 72.9 pg/ml. They were placed on a six-month discontinuation (drug holiday) from the bisphosphonate with their CTX values then ranging from 162 pg/ml to 343 pg/ml with a mean of 228.2 pg/ml. The mean monthly improvement was approximately 25.9 pg/ml per month. Five of the 17 patients needed surgery to resolve their problem, with 2 needing extensive mandibular resection (removal of a section of the lower jaw). Below is a summary of information for the 17 patients.

Patients taking oral bisphosphonates when first seen:

Patient

Years of bisphosphonate use

CTX pg/ml

CTX after 6 months drug holiday

CTX change

4

6.4

102

291

189

7

8.4

102

199

97

10

7.3

53

343

290

11

5.5

88

299

211

15

6.3

64

204

140

20

9.1

79

302

223

24

10.2

33

202

169

28

9.3

66

232

166

19

7.0

99

212

111

14

3.3

30

175

145

22

8.8

90

162

72

26

5.0

94

216

122

2

4.4

72

188

116

12

4.9

53

194

141

21

5.0

101

303

202

17

3.3

80

196

116

13

8.6

33

162

129

Eleven patients had resolution of their osteonecrosis without surgery. The mean CTX value for this group was 245.8 pg/ml after a drug holiday of 3 to 9 months. There were 13 patients that had discontinued bisphosphonates at the time of their first visit. Their CTX values ranged from 153 pg/ml to 303 pg/ml and correlated with the length of time on the drug holiday. Their detailed information is presented below.

Patients on drug holiday at time of first visit.

Patient

Drug holiday months

CTX pg/ml

1

2

153

3

6

244

5

4

206

6

12

294

8

2

162

13

10

303

16

8

249

18

3

189

23

7

216

25

8

232

27

4

194

29

15

303

30

6

199

Data analysis. Although there is no exact number for individuals taking oral bisphosphonates, the authors suggest that it is certainly in the tens of millions. Thus showing the rarity of BIONJ since there were only 30 cases in this one medical center. The authors propose several possible reasons for their being such a high percentage of BIONJ cases in this study that were taking Fosmax compared to Actonel. But eventually conclude: "Therefore, the difference in the incidence of osteonecrosis between the 3 oral bisphosphonates in use today relates to its accumulation and retention in bone and a clear explanation as to why there are differences may never be completely known."

Duration of bisphosphonate use and its importance. The time during which the drug was taken, and thus allowed to accumulate in bone, is clearly important. No one taking a bisphosphonate less than 3 years developed BIONJ in this study. The authors note the following important information, too: "Moreover, the data indicated that most cases developed after 5 years or more of drug exposure and that the incidence and severity increased linearly with exposures over 3 years." They note this differs from what happens with IV bisphosphonates where BIONJ develops in only a mean of 9.3 months for Zometa and 14.1 for Aredia. They suggest this is because the IV method bypasses the low and variable gut absorption of oral bisphosphonates.

Gender relationship of BIONJ. The fact that all patients in this study were women is not unexpected since this is the biggest group of individuals taking oral bisphosphonates. Even regarding BIONJ in IV bisphosphonate use, there is a 70:30% bias toward women.

Why is the mandible more prone to BIONJ? Bone is constantly aging and needing to be replaced. The osteoclasts are the cells that resorb the old bone which is replaced by osteoblasts. The bisphosphonates block the osteoclasts from breaking down bone. Bone that turns over more rapidly would thus be more greatly affected by the bisphosphonates. The authors note, "Therefore, the rate of bone turnover in any bone directly relates to its vulnerability to bisphosphonates. The mandibular alveolar bone turnover rate is 10 times that of long bones like the tibia." They also note that the rate of bone turnover at the crest is twice the rate of bone turnover at the mandibular canal and 3 to 5 times that of the bone about the lower border. The region where bone turnover is the highest is exactly where BIONJ occurs most frequently. (Not mentioned but potentially important is that the crestal mandibular bone is also most subject to trauma from chewing, brushing, etc., thus making it more prone to exposing the bone under the thin soft tissue, a potential cause of BIONJ).

Pain and other considerations. Pain correlated directly to evidence of infection since nonvital bone is deinnervated and nonpainful. The pain comes from secondary sources that invoke the mediators of inflammation. The authors noted, "The size of the bone exposure, pain and infection, and the requirement to accomplish surgery all directly correlated to the length of time the patient had been taking the oral bisphosphonate."

More on the CTX usage. The morning fasting serum CTX test measures a specific crosslink peptide found in type I collagen, which accounts for 98% of the total protein in bone. Serum CTX is considered to be the best indirect measurement of bone turnover available and it is performed in only one laboratory in the United Stated. The normal values are reported out as 50 pg/ml to 450 pg/ml, however, the authors note these don't apply to individuals treated for osteoporosis. Regarding the values for individuals not being treated for osteoporosis, they state, "Actual normal values are usually well over 300 pg/ml and are most commonly 400 pg/ml to 550 pg/ml." Thus the lower values seen in individuals treated for osteoporosis show some degree of suppression of bone turnover.

The 17 patients who were initially taking a bisphosphonate and who were then placed on a drug holiday provide the best data set to evaluate the degree of suppression and the effects of the drug holiday. All these patients had an initial value of less than 110 pg/ml. There is an apparent 25 pg/ml day-to-day variability in the CTX test, but the authors note that even with that, all would have scores less than 150 pg/ml. When placed on a drug holiday, the average six month improvement was 155.3 pg/ml, or about 25.9 pg/ml per month improvement. This correlates well with the 26.4 pg/ml per month improvement in the patients who were already on drug holiday when first seen. Thus meaning this monthly rate of increase should be a valuable tool to predict improvement in bone turnover for future patients. Successful treatment of BIONJ is dependent upon improved rates of bone turnover, which don't happen when the IV bisphosphonates are stopped. This appears to be due to "exhaustion atrophy" of the osteoclast precursor cells in the bone marrow due to the extreme effectiveness of IV medications at blocking bone turnover.

Interesting observations from studies where Fosamax was stopped after some years of therapy. Here is one quote with interesting implications, "It is important to note that in a 10-year study by Bone and Hosking et al on osteoporosis patients treated with Fosamax, the BMD values improved to a maximum at 3 years and that the group that stopped taking Fosamax at that point experienced no increase in fractures compared with the group that continued taking Fosamax and that their BMD values decreased only slightly over several years. A confirmatory study by Black et al noted in a randomized double-blind controlled study of 760 women that once Fosamax had been taken for 5 years, no increase in fractures occurred after another 5 years off the drug." [Emphasis is from the editor and not in the original article]. Note that we measure osteoporosis and success in treating it by measuring BMD with dual-energy X-ray absorptiometry (DXA), but in reality it is fracture prevention that is most important.

Comments on the drug holiday. The authors note, "These data combined with the CTX data presented in this report indicate that drug holidays on the order of 4 to 6 months are reasonable, safe, and could be expected to minimize the risk of osteonecrosis (due to the anticipated rise in CTX due to increased osteoclast function) when performing invasive oral surgical procedures in patients taking oral bisphosphonates while also maintaining BMD values and fracture prevention related to the osteoporosis."

Prevention and treatment recommendations (summarized), clinicians should get a copy of the original article to get all the details.

Prevention. Dental clinicians should use the first 3 years after a person starts bisphosphonates to provide any needed dental work since the individual is at the lowest risk for BIONJ at that point. This therapy should include a thorough medical and dental history and oral examination. Oral surgery should be performed first, followed by periodontal therapy, then endodontics and restorative dentistry. Little is known about the effect of orthodontics, but it is presumed safe in the first three years, as are implants also expected to be safe, effective and able to integrate during that time period.

Patients seeking treatment already on a bisphosphonate. The dentist should ask how long the patient has been on a bisphosphonate and if they are also being treated with corticosteroids or other medications that might affect bone healing. Supragingival therapies should not affect BIONJ during any time period since the patients started bisphosphonates. For patients taking bisphosphonates fewer than three years, bone healing should be normal if surgery is needed. CTX should not be needed but can be taken at that time to assess the degree of bone turnover suppression and/or to establish a baseline reading. With greater than three years bisphosphonate therapy, or fewer than 3 years with concomitant corticosteroid therapy or chemotherapy, a CTX is highly recommended. If the CTX is 150 pg/ml or greater, then invasive oral surgery can probably be accomplished safely. If the CTX reading is less than 150 pg/ml, then contact the physician to discuss the possibility of a drug holiday with or without substituting other FDA-approved osteoporosis medications before initiating surgical procedures. Other medications that could be substituted include raloxifene (Evista, for women only), calcitonin salmon (Miacalcin) and recombinant human 1-34 parathyroid hormone (Forteo). After a 4-6 month drug holiday repeat the CTX and extend the holiday if the reading is less than 150 pg/ml with the agreement of the patient's physician. In the rare event that the physician doesn't want to institute a drug holiday, it is best not to consider doing elective procedures. If urgent invasive procedures are needed, inform the patient of the added risk of BIONJ and do the surgery.

Treatment of BIONJ. Treatment of BIONJ is usually less extensive and more responsive to treatment than IV bisphosphonate-induced osteonecrosis of the jaws. Contact the physician and try to correlate therapy with the CTX value and efforts to reduce it via a drug holiday and/or adding a non-bisphosphonate medication to the therapy. If the exposed bone is painless, start the patient on 0.12% chlorhexidine. If pain is present, in addition to the 0.12% chorhexidine, use Penicillin V-K 500 mg 4 times daily for people not allergic to penicillin or use levoflaxacin (Levaquin) 500 mg once daily if they are allergic to penicillin. Next choices include doxycycline (Vibramycin) 100 mg once daily and zithromycin (Zithromax) 250 mg once daily. In patients refractory to antibiotics metronidazole (Flagyl) can be added using 500 mg 3 times daily. Clindamycin is not recommended due to its ineffectiveness on the microorganisms involved. Use the antibiotics for 14 days or until pain is controlled or returns. The oral surgeon should resist the temptation to debride locally inititally since many cases of BIONJ will heal spontaneously during the drug holiday. If exposed bone becomes mobile or there is radiographic evidence of a sequestrum, that can be debrided locally with expected healing if the CTX value is greater than 150 pg/ml.

Summary. The comorbidities of prednisone and/or methotrexate don't cause BIONJ, but they cause it to occur sooner and to be more severe and to make it respond more slowly to a drug holiday. Interpret the CTX as follows: if the value is 100 pg/ml or less this is high risk,100-150 pg/ml is moderate risk, and greater than 150 pg/ml is minimal risk. Use these values to guide treatment decisions. It is possible to provide comprehensive care to patients treated for osteoporosis with bisphosphonates if dentists will follow the guidelines in this article.

Editor's comments: This article provides a lot of guidance to patients taking oral bisphosphonates. It is important to be proactive and to discuss this information with your physician and/or dentist, especially if you plan on having any surgical dental procedures done, including dental implants. In that case you would definitely want to know what your CTX value is so you can make informed decisions about the type of dental therapy to have. Items of importance:

1. Non-invasive dental therapy should not be an issue no matter what your CTX score is, so there is no need to have that test done for normal non-surgical procedures such as X-rays, examinations, tooth cleanings, etc.

2. Be sure to see a dentist for an exam should you develop any lesion involving the gums or soft tissue that exposes bone, especially of your lower jaw (mandible). Even though this is painless, it could be BIONJ and would then need to be followed and/or treated based upon your CTX value.

3. This article doesn't mention it, but I could imagine an ideal situation where you have a fasting serum CTX before you start taking a bisphosphonate to treat your osteoporosis for a baseline reading, then have a follow up done at about 2.5 years. If that test shows bone turnover to be severely depressed (less than 100 pg/ml), then you might see about a reduced dose of the bisphosphonate or occasional short drug holidays to reduce the risk of BIONJ by keeping the CTX value at or above 150 pg/ml. This would be similar to taking thyroid medications where the drug is titrated to be sure you are getting the ideal response and adjusting the dosage according to the blood test results.

4. Since individuals who have taken bisphosphonates for longer periods are at greater risk of more severe BIONJ, it appears appropriate to ask for a fasting serum CTX. Then if your CTX value shows severe bone turnover suppression, you might also want to ask about a drug holiday or to be placed on another medication that works differently. Note that there appears to be little chance of increased fracture risk even after stopping Fosamax for five years, so a year or two drug holiday might be an option, too, especially if you have reached normal BMD while on the bisphosphonate. But you would want occasional CTX testing and/or DXA done to evaluate your response to the drug holiday.

5. This article discusses only oral bisphosphonates for treatment of osteoporosis, but there are now IV bisphosphonates such as Boniva being used for that purpose too. All the precautions and suggestions should apply equally no matter what form of bisphosphonate you are taking for osteoporosis, oral or IV.

6. It is interesting to speculate why there were so many more cases of BIONJ with Fosamax than Actonel. One obvious thing is that Fosamax has been around longer and was considered the standard for quite some time, thus there are perhaps many more people taking Fosamax. I don't know how to get figures for the percentage of the population that uses each medication although it would be insightful I'm sure. One might think that because Actonel is taken as a 35 mg pill once weekly and Fosamax as 70 mg once weekly, that might explain it. However, my readings indicate that Actonel is about 10 times more potent that Fosamax, thus why the reduced dosage works equally as well. For now we must await further studies to see if there is truly any risk reduction by taking Actonel vs. Fosamax. Since they both work exactly the same way it is hard to imagine why there would be a significant difference at equivalent doses.

Return to Home