Some manufacturers have reported a possible correlation between the use of resin-modified glass ionomer, or RMGI, cements and the premature fracture of all-ceramic crown restorations resulting from hygroscopic expansion of the cement.

Because of practitioners’ concerns regarding the efficacy of cementing all-ceramic crowns with RMGI cements, we undertook two pilot studies. The purpose of these studies was twofold: first, to test the Procera AllCeram aluminum oxide coping material for fracture due to expansion of the cement when luted with two of the more popular RMGI cements, and, second, to determine if continued expansion over time may weaken the restoration, causing premature fracture.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
To conduct these studies, we obtained grade 2 titanium standardized dies, premilled to the dimensions of a mandibular molar crown. One of us (M.S.) scanned the dies using computer-aided design/computer-aided manufacture technology. We then forwarded the data to a laboratory where the sintered aluminum oxide copings were fabricated. On their return to us, we visually inspected the copings and matched them to their specific titanium die. The dies and copings then were divided into their respective groups. We mixed the cements according to the manufacturers’ instructions and placed them in the copings. We then luted the copings to the titanium dies, removed the excess cement and placed the cemented copings under a load of 11 pounds of force for 10 minutes to ensure proper seating. The copings were placed in 100 percent humidity at room temperature until the time of testing.

The study results showed that the mean load to fracture was significantly greater in samples that were tested 60 weeks after cementation.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
In the first study, we divided 30 all-ceramic sintered aluminum oxide copings into two groups of 15 each. Two RMGI cements (Vitremer Luting cement [now RelyX Luting Cement], 3M ESPE, St. Paul, Minn., and GC Fuji Plus, GC America, Alsip, Ill.) were used as the luting agents. We cemented the copings to titanium dies and placed them in a 100-percent humidity chamber. We inspected the cemented copings at six, 12 and 24 weeks using x2.50-magnification surgical loupes, as well as under x20 microscopic magnification.

On inspection at six weeks and 12 weeks, we found no signs of fracture in any of the 30 samples. However, at 24 weeks, we found that one coping cemented with GC Fuji Plus had fractured vertically on both the mesial and distal sides (but the fracture did not cause the coping to separate from the titanium die). No other GC Fuji Plus samples and none of the Vitremer samples exhibited any signs of fracture at the 24-week inspection. Seventeen additional samples that were cemented at the same time as the 30 test samples have remained in 100 percent humidity; we inspected them four years after cementation. They exhibited no visible fractures when viewed with x2.50-magnification surgical loupes and under x20 microscopic magnification.

Continued research by dental product manufacturers has shown that fracture rarely, if ever, is due to expansion of the cement alone. Moreover, manufacturers have modified the composition of RMGI cements so that expansion has been reduced to minimal amounts.

The possibility remains that visibly undetectable microfractures might develop in all-ceramic crowns as a result of continued post-cementation expansion of resin-modified cements. If these microfractures do occur, they might lead to continued weakening and eventual premature fracture of the crown restoration under normal occlusal forces. To investigate this possibility, we designed and conducted a follow-up pilot study.

In the follow-up study conducted one year later, we divided 18 aluminum oxide copings into three groups of six samples each. Each group of six samples was cemented with GC Fuji Plus one week before testing, 10 weeks before testing or 60 weeks before testing. Just before testing, we inspected all 18 samples under x20 microscopic magnification and found no detectable fractures in any of the samples. Using a universal testing machine, we then loaded the samples to fracture at 0.5 millimeter/minute. As illustrated in Table 1, the mean load-to-fracture value was significantly greater in the samples cemented 60 weeks before testing than those cemented one week before testing.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

In 1998, Procera introduced its AllCeram system, which incorporated a sintered aluminum oxide coping veneered with aluminus porcelain to create a strong and esthetic all-ceramic crown. This was a significant advancement because it allowed the practitioner to use his or her favorite luting agent without the need to be concerned about fracture under occlusal load. However, the ability to use any RMGI cement introduced a concern that the expansion of these cements could cause the all-ceramic crowns to fracture because of the continued postcementation expansion of the luting agent.

The study results show that fracture of the all-ceramic copings resulting from expansion of the cement alone did not occur.

To date, some manufacturers still do not recommend the use of RMGI cements for luting all-ceramic crown restorations; however, our studies have shown that the expansion does not cause fracture of the all-ceramic crowns, nor does the potential for continued expansion weaken the crown. In fact, the load-to-fracture force was significantly higher the longer the coping was in place.

Craig and Powers⁴ reported that the average biting force is 665 newtons (or approximately 150 pounds of force) for natural teeth in the molar region. The results of our pilot studies showed that after cementation, the load to fracture was well above the average occlusal force applied in the molar region; moreover, as time passed, the force required to fracture the samples increased dramatically, providing additional resistance to fracture for the underlying support system.

We recognized that by applying the force in a cyclic manner (to simulate chewing), a testing condition more similar to clinical conditions might have been created; however, these studies used a static load (that is, applying force at a continual, slow rate) to measure the maximum force required to fracture the underlying support structure. We are planning further studies that will introduce new variables—such as adding veneering porcelain to the coping and cycling the samples—to determine how these variables affect the all-ceramic system.

	CONCLUSIONS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The results of our studies show that fracture of the all-ceramic copings resulting from expansion of the cement alone did not occur. The mean load-to-fracture values were significantly greater for the samples tested 60 weeks after cementation than those for samples tested one week or 10 weeks after cementation. Using the load-to-fracture value as an indicator of success, we found that the copings were not weakened over time as a result of potential microfractures, but that the compressive strength was significantly greater in the samples that were cemented 60 weeks before undergoing testing.

The clinical implications are evident considering the increasing use of all-ceramic restorations throughout dentistry. The advantages include excellent esthetics, use of a strong coping system and flexibility in choosing a cement.