JCO Interviews Robert J. Schulhof on Functional Appliance Results
GOTTLIEB Bob, in your article in the September issue of JCO entitled "Results of Class II Functional Appliance Treatment", you presented an analysis of the results of treatment with four functional appliances. We have had a variety of responses to the article that I think we ought to discuss, but especially the apparent deficiency of the Frankel appliance, since it was the only one of the four not to show more than expected normal growth.
SCHULHOF I wouldn't want to say that the article was a comparison of the various functional appliances and what they do. It is more a question of what is the potential of functional appliances in general. I wouldn't want to say that any of those samples was so pure that we knew exactly how bionators compare to Frankels, or that we might not have gotten a different result if we had someone else's bionators and someone else's Frankels. While there are some lessons we can learn from the paper, it would not be that an activator does more than a bionator, and a Frankel does less than either of them. I don't think that is valid, since there was only one operator per appliance. So, we don't know whether we are measuring the difference between appliances or the difference between operators.
GOTTLIEB Another operator might conceivably perform differently with the same appliance or something that is called the same appliance. That is a problem we have in talking about functional appliances-- the importance of knowing what appliance design was involved-- because appliance design might make a difference. Dr. Frankel is reported to have complained about improper pad placement, improper vestibular shield design, improper clasping, lack of notching of the teeth, and the position of the lingual wire. He apparently feels that appliance design with a Frankel may have a great deal to do with the result you can get. Apart from appliance design, there are a number of factors that might conceivably be involved with variations in results with functional appliances. We might mention such additional factors as age, case selection, cooperation, the severity of the malocclusion, the amount the mandible is advanced, facial pattern, and individual response.
SCHULHOF Case selection and facial pattern are Number One, cooperation is Number Two; but all of those factors may be significant to the result. In our study, cooperation was adequate, because the Class II was corrected, and the sample sizes were large enough to average out individual response. There may be many other factors that can account for differences in results, and they are still open to further study.
GOTTLIEB In the article, the Frankel sample differed from the rest in being the only one not to show an average additional growth of the mandible. The sample only grew to the extent that was normally expected. Could this be an aberration caused by a defective sample?
SCHULHOF It could be. However, we have had two other Frankel samples that showed the same result. I still have not found a Frankel sample that grew more than in proportion. That doesn't mean that there will not be one. I hope there will be. Growth prediction technology is complicated enough without having to have a separate prediction formula for each functional appliance.
GOTTLIEB A comment has been made that you were comparing select samples to the average, since it might be expected that the samples used had some selectivity in the choice of cases to be treated and in the further choice of the cases to be included in the sample. Since they were successfully treated functional appliance cases, maybe they grew more than average in order to be successful and in order to be in your sample, and maybe that's the reason that the results showed more growth on the average than normally expected.
SCHULHOF To answer the first part, case selection is a thread that runs through functional appliance therapy, and I would not under-emphasize the benefits of selecting cases with good growth potential in amount and direction for treatment with functional appliances. The answer to the second question-- whether, since these were successfully treated functional appliance cases, they grew more than average and that is the reason they were successful-- is twofold. First of all, the Frankel cases were treated successfully-- the Class II was corrected-- but they did not grow more than average. Second, I would like to refer to the work of Reey (AJO, April 1978) in which 50 consecutively treated activator cases had exactly the same average result.
GOTTLIEB Meaning the 2-3mm average additional growth in the mandible?
SCHULHOF Yes. So we do have one sample that we call pure in selection. They were brachyfacial. They were selected according to previous knowledge of what would succeed, but they were not selected after having been successful. This was a prospective study.
GOTTLIEB Might not the selectivity just produce a sample of good growers?
SCHULHOF We did take a sample of good growers that matched the functional appliance samples closely, but were treated with fixed appliances, and compared their results with the growth forecast. Those people did not grow more in the mandible than the growth forecast indicated .
GOTTLIEB Could one not question both the matching and the forecasting?
SCHULHOF If this were the first time we had ever used the growth forecast to compare a sample of treated cases, then we might be a little bit uncertain, but this is probably the hundredth time we have used this tool and found it to work. The only time we have ever found a significant difference between average mandibular growth and a sample has been in samples of Class Ills and one sample of Class II division 2s that had more than a 150° interincisal angle and more than a 6mm overbite. They grew more in the mandible compared to the cranial base than the growth forecast, and we have since had to modify our growth forecast for certain types of Class III.
GOTTLIEB There is objection to using the condyle axis and the corpus axis to assess growth of the mandible, because they depend on the proper location of Xi point and because the condyle axis ends at basion-nasion. If the condyle moves downward and forward, the line drawn from Xi point to basion-nasion lengthens and it would seem as though there had been condylar growth even if there had not been.
SCHULHOF That is a valid objection, and I would suspect any single cephalometric measurement, because there are so many subtle changes that are occurring simultaneously. We used the ratio of condyle axis and corpus axis to cranial base for purposes of presenting statistics in the article, but to make sure that we were not misled, we created complete composites of the samples by digitizing 100 points, 20 of them on the mandible alone.
GOTTLIEB How were the growth forecast diagrams made?SCHULHOF There are two ways to do a prediction. You can tell the machine to predict an entire face with the amount of growth you would expect to get in a certain number of months, or you can request a prediction based on the actual amount of growth that there was in the cranial base.
GOTTLIEB The second one is a better way of doing it, isn't it?SCHULHOF That's the way we did it in the study. Using that method and digitizing 100 points to form the composites, you can lay mandible over mandible and see everything that happened. That'sreally the only method that shows what is going on, and it is far superior to any single measurement. For example, I wouldn't want to use lower incisor to APo to define incisor tipping, because A and Po are doing different things. If you use facial axis or facial angle for chin position, it doesn't explain how it got there.
GOTTLIEB But if you wanted to assess change in chin position relative to the rest of the face, you would use facial angle.SCHULHOF Absolutely. That would tell you want kind of result you got. It doesn't tell you it grew. It tells you it changed. To study the mechanism of growth, the digitized composites are a method of choice.
GOTTLIEB How do you digitize the condyle? It is not easy to find on all cephalometric x-rays.SCHULHOF There is more error in finding the condyle, but the errors average out over an adequate sample. Samples are made to average out errors of that kind. The technicians who digitize these have no reason to be biased one way or the other, and, therefore, the errors would be totally random and average out.
GOTTLIEB You did find that the ratios of change in condylar axis and corpus axis relative to BaNa were statistically significant.SCHULHOF Yes.
GOTTLIEB You also found a range in which some are remarkably different.SCHULHOF There were two very interesting things about the functional appliance results. One was that the average total mandibular growth from the condyle to the symphisis averaged 2-3mmmore than would have been expected if the cases had not been treated. However, the range was 0-6mm more. It was fascinating that not only did the mean change, but the standard deviation changed disproportionately.
GOTTLIEB You have much bigger standard deviations with functional appliances.SCHULHOF Right, and this may cause orthodontists to get into trouble, because everyone who uses functional appliances is going to find a few superstars among his treatment results, and those are the ones that are going to be shown at meetings. Others may assume that they can expect that result every time, and they cannot. Absolutely incredible results do occur, but to represent that as what will occur in the majority of cases is not true.
GOTTLIEB One of the criticisms of the whole functional appliance movement is that you cannot predict who will respond and who will not, and, therefore, you might spend a couple of years of someone's time in a fruitless effort.SCHULHOF The Class II Frankel cases did correct to a Class I occlusion, in spite of the fact that the cases in the sample on average did not grow more than they would have grown without treatment. But, the results looked pretty good. I liked the composite.
GOTTLIEB So, something else happened.SCHULHOF It means that growth isn't the whole story and that, in many cases, you can perform a pretty decent treatment even if you don't get extra growth. In a brachyfacial patient who has a 95° facial axis, you don't have to get super growth. Any growth you get is going to be helpful to his treatment. On the other hand, in a dolichofacial patient with an 85° facial axis you have to have extra good horizontal growth in order for the treatment to work. That may be why we see such a bias toward brachyfacial cases in the treatment results presented. Those are pretty good facial patterns and, even if you don't get growth exceeding the expectation for that individual, the odds are still with you to be able to complete the case.
GOTTLIEB One criticism of functional appliance therapy by fixed appliance orthodontists is that easy cases are chosen to be treated with functional appliances, while fixed appliances treat the "real" Class IIs. Do functional appliances treat the "real" Class IIs?SCHULHOF If you were to ask an orthodontist what he considers to be his most difficult cases to treat, he would say open bites and dolichofacial Class IIs with retrognathic mandibles. We have no indication so far that functional appliances help to treat many of those cases. If you happen to be the one out of ten people who gets 6mm more of horizontal mandibular growth, you may have saved yourself a surgical procedure; but that is not going to happen in a majority of cases, even though it is an interesting potential. However, I'd like to leave that door open. I do have reports from somesources that claim successful treatment of dolichofacial patterns with functional appliances. I haven't gotten sufficient data yet to make any comment on that. I would be very happy to get more data on dolichofacial patterns treated with functional appliances. I would certainly be willing to work with anybody who has such data.
GOTTLIEB How do you respond to the orthodontist who says that possibly 20% of the average practice might be suitable for functional appliances, that you may get 50% of the kids to wear one, and that of those who wear them, one in ten-- or 10%-- may get the outstanding result? Ten percent of 50% of 20% is 1%.
SCHULHOF The percentage suitable for functional appliances is probably more like 40%. The appliance will give acceptable results on almost any case on the normal to brachyfacial side where the maxilla does not require modification. With regard to cooperation, if you are trying to sell anything, the extent to which you believe in it has an awful lot to do with your success rate. However, there is no question that these appliances require excellent cooperation. I would have to agree that it may be one in ten who gets the extraordinary effect. However, on the 40% you can count on a fairly decent result, if you get proper cooperation, although to achieve an ideal result you may frequently have to finish with fixed appliances.
GOTTLIEB Of this 40%, what percentage could not be treated just as well with fixed appliances? SCHULHOF Those brachyfacials who are one standard deviation or beyond in low mandibular plane angle or short lower face height would not be as successfully treated with fixed appliances, because you have to intrude incisors a long way. Those patients come out better esthetically with a lower face height closer to normal. These abnormal brachyfacials probably constitute 20% of all cases.
GOTTLIEB What do functional appliances do even for brachyfacial patterns that can't be done as well and faster with a fixed appliance? SCHULHOF If you have to resort to bite-opening mechanics rather than just intrusion mechanics, you are probably better off with a functional appliance. With Class II elastics, you rotate the mandible open, and this causes the jaw to go down and back when you wanted it to come forward. You benefit in the vertical and lose in the horizontal. With functional appliances, you can get the benefit in both directions on that case. The one functional appliance treatment that will succeed every time is on the brachyfacial case with inadequate lower face height. The functional appliance will consistently increase their vertical dimension and make them look more normal. Increasing the vertical in a deep bite case with a functional appliance is probably going to be more stable than if it were done with a fixed appliance, since it is accompanied by condylar growth. However, if you don't control the inclination of the incisors by following up with a fixed appliance to finish treatment, that may relapse. We did a study on deep bites and found that the higher the interincisal angle, the greater the deep bite relapse. So, if you are just using a functional appliance and not controlling incisor inclination, you are probably leaving the case open for relapse even if you are intruding a lot. The functional appliance minimizes the amount of intrusion needed and creates a stable vertical increase, because it is accompanied by additional condylar growth. If you follow with a fixed appliance, you can get the right incisor inclination. The combination of functional and fixed appliances would be indicated for many patients. GOTTLIEB Bob, I'd like to summarize what we have been saying. The article was an effort to present an analysis of four functional appliance samples. While it may invite comparisons, it was not intended to compare these samples to the advantage or disadvantage of one appliance or another. Other samples for the same appliances might show different results. There are many variables to be considered, such as different aspects of the samples and the operators. Although functional appliances do have a potential for growing mandibles an average of 2-3mm more than normally expected growth, the standard deviations are high, which means that you can occasionally
see extraordinary results. From what we have observed so far, functional appliances seem to work better on deep bite brachyfacial Class II cases. However, their potential on other cases, including dolichofacial Class IIs with retrognathic mandibles, remains to be examined. Of the 40% of brachyfacial Class II cases in the average practice, the 20% that are deep bite, short vertical face height cases probably are best treated with a combination of functional and fixed appliance therapy, since the functional appliances have the potential to result in a more stable increase in vertical dimension, but have limitations in putting the teeth at desired angulations. From what we know so far, it would appear that there is a place for functional appliances in orthodontics, but that success is dependent upon discrimination in case selection, motivation in patient cooperation, and attention to individual response-- and a variety of other factors that remain to be investigated.
