Salvaging a failing case

Published: December 2011

Bulletin #6 - December 2011

Salvaging a failing case

If you set yourself up as an expert in a small corner of any specialty, then you must expect to receive a good number of patients from other co-specialists who prefer to deal with other small corners of your shared profession. These are the cases that you will welcome because, with your extra knowledge and experience of their condition in your small corner, you have “what it takes” to treat them better than most and to really establish your reputation. On the other side of the coin, however, are those patients whose treatment was begun by a colleague who, not initially recognizing the complexity of the problem, accepted them for treatment for the condition which he/she was ill-equipped to undertake. The progress of the treatment grinds to a halt facing potential failure and, with no therapeutic answer to hand, the specialist concerned “dumps” the patient on you.

In these latter cases, the treatment may have gone too far in the wrong direction for you to offer any salvation. For some of them, you may be able to back pedal a little, re-direct the treatment and produce an acceptable outcome. With impacted teeth, this almost invariably means settling for a compromised result, one which may produce a successful resolution of the impaction but which may additionally involve collateral damage to the surrounding structures, which could have been avoided had the patient been referred to you earlier.

These cases are the failed, partially failed or eeked-out-result cases and they deserve to be analysed as to what went wrong, why it went wrong, what was done to turn the case around and to what degree the outcome was compromised. It will be seen that there are such a large number and range of operative variables involved, from both the orthodontic and surgical aspects, that evidence-based recommendations are hard to come by and clinical experience and common sense must reign supreme. Nevertheless, cases of this type provide much material for discussion for what we may learn from their management and from a study of the clinical decisions that were made during the treatment.

For the present December 2011 bulletin, therefore, I shall present a single case which displayed a series of potential obstacles to the achievement of a successful outcome, some of which were recognized and planned for, in advance …………. but not all. The case was heading for failure with the distinct possibility that each of the anterior 6 teeth could be lost.


Fig. 1. Panoramic view November 2003 shows the canines close to the apices of the lateral incisors.

Fig. 2. from the cephalogram December 2003 the crowns of the superimposed R & L canines, in this true lateral view, are in the same line as the long axes of the superimposed R & L central incisors. The outline profile of the lateral incisors can be distinguished, with the apices palatal to the canine crown tips. Thus, the canines are labial!

Fig. 3. periapical view November 2006 shows the auxiliary labial arch ligated to the impacted canine and the curved roots of the distally tipped first premolar obstructing the ridge area.

Case report:

The patient was an 11 year old female first seen in December 2003, when initial records were taken. She had a late mixed dentition, with mobile or recently exfoliated deciduous second molars, but with maxillary deciduous canines which were firm and showed no signs of imminent shedding. There were no clinically palpable signs of the unerupted maxillary permanent canines on either side of the ridge. Each maxillary first premolar exhibited a mild rotation of the palatal cusp to the mesial. The maxillary lateral incisors were tipped slightly labially in comparison with the upright central incisors. The molar relations were a half cusp width (cusp-to-cusp) class 2, the overjet normal and there was an open bite tendency in the incisor region, with occlusion only on the permanent molars. Sufficient space for all the permanent teeth appeared to be present in both jaws.

The panoramic radiograph showed all permanent teeth developing, with the exception of the maxillary third molars. The mandibular third molars were situated at a 90 degree angle in the ascending ramus on each side. The unerupted premolars and second permanent molars appeared to be normally developed in relation to her chronologic age, but the maxillary deciduous canines showed unresorbed roots and their permanent successors were seen to be close to the apices of the lateral incisors, with an enlarged dental follicle on the right side. The mesio-buccally rotated first premolars displayed a markedly distorted palatal root which curled from the vertical to the horizontal, in a palatal direction. The film showed closed central incisor roots, although the lateral incisors did not seem to have completed their apexification and the roots of all four incisors appeared to be fairly short. The orthodontist considered that these had already suffered some apical resorption.

The lateral cephalogram was taken at approximately the same time and was prescribed for the purposes of overall orthodontic diagnosis and treatment planning. The details of the cephalometric analysis can be safely put aside for the present discussion. What is pertinent, however, is that from this true lateral aspect and from the outline of the superimposed unerupted canines and of the incisors, the canine crowns are seen to be in a direct line with the central incisors and slightly labial to the root apices of the lateral incisors. This accounts for the lingual displacement of the root apices of the lateral incisors, clearly seen on the cephalogram and for the consequent labial tipping of their crowns, noted on the clinical photographs.

A CBCT examination was not carried out at that point in time because, in Israel in 2003, CBCT was only in its early stages of professional acceptance.

The Initial Phase of Treatment:

Orthodontic treatment with the use of a bonded Edgewise appliance began in February 2004 in the maxilla only, with the aim of opening space for the canines in the dental arch. In July 2005, the patient was referred for the surgical exposure of the canines, which was performed under local anesthetic by an oral and maxillofacial surgeon, using a palatal approach. The canines were very high up and close to the midline on each side. The orthodontist was on hand to bond two eyelets with a twisted ligature on each to the minimally-exposed and anatomically-palatal aspect of each canine. No other surface of the canines would have been accessible from the palatal side without excising a liberal and damaging quantity of bone and follicular tissue, not to mention unnecessarily exposing much of the roots of the incisors. During the bonding procedure, the surgeon maintained the hemostasis and isolation of the two teeth, with the aid of retractors and high powered suction. Following this, the surgeon sutured back the flap into its former place, taking care to pass the twisted ligatures from the eyelets through the palatal flap at locations immediately opposite the two canines. The twisted steel ligature wires that had been threaded through the bonded canine eyelets were shortened and curled into a hook, to lie passively close to the replaced palatal flap.

At the completion of the surgical episode, the surgeon considered it advisable to perform a CBCT, in order to monitor the root resorption and because of the degree of vertical displacement of the tooth and the difficulty of access. The patient was referred to an imaging institute for this to be done. The CT views showed a greater degree of root resorption than had been present on the initial records.

The patient returned to the orthodontist only 2 months later, for him to place a “ballista” spring1 on each side in the auxiliary molar tubes, in addition to the heavy main arch. In the canine area he turned the vertical portion horizontally across the palate and against the palatal mucosa where they were held in place by the twisted ligatures that were bent around the terminal loops of the ballista. The ballistas were thus applying extrusive force to the canines, through the agency of the twisted ligatures. The patient was seen on a regular basis thereafter by the orthodontist who periodically re-approximated the loops against the palatal mucosa, as they became distanced, following the vertical response of the canines to the downward-directed extrusive forces. After 4 months of vertical traction, elastic traction was applied from the twisted ligatures in a horizontal and posterior direction to the cleat on the first molar bands of each side.

Two months later, the canines were seen to be sub-mucosally bulging the palatal tissue, close to the midline, in the mid-palate. The overlying mucosa was therefore removed and the teeth allowed to erupt further. The canines erupted with their anatomically-labial surfaces facing each other on either side of the median raphe, in mid-palate. At this stage, new radiographs were ordered and these revealed considerably more resorption of the roots of the four maxillary incisors.

The parents were very concerned, sought my advice and subsequently requested that I assume responsibility for the continuation of treatment.


Figs. 4. The clinical views of the teeth in occlusion, December 2006 at the time of transfer. Note the large overjet, adequate space in the dental arch for the canines, broad open bite. Note also the incorrect positioning of the bracket on the left first premolar and the consequent distal crown tip. The occlusal view of the maxillary arch, with the erupted canines rotated 180 degrees and the eyelet attachments on the anatomic palatal surface. Note that in this view, the height differential between the canines in the vault of the palate and the labial archwire cannot be appreciated.

I first saw the patient in December 2006, 17 months after the surgery and a little under 3 years since the commencement of treatment. The was a very broad open bite from first molar to first molar and a large horizontal incisor overjet. Adequate space for the canines was present.

Aside from the severe resorption, a study of those latest radiographs revealed abnormal anatomy of the palatal roots of the first premolar on each side, which were curled palatally and mesially such that their apices were in the direct line of the canines to the spaces prepared for them in the arch. The bracket on the left first premolar had been poorly sited and had generated a distal crown tip during the earlier space-opening procedure.

The bottom line in this case was that, if the continuation of the treatment was performed with insufficient attention to detail, then there was a real risk of the child losing the maxillary 6 anterior teeth!

The dilemmas at that time:

  1. The canines were facing one another in the mid-palate, rotated 180 degrees. Was it reasonable to attempt to align these teeth in the arch from that distance, to rotate them, torque them and expect a good prognosis? Perhaps they should be drawn to the arch in their rotated condition?
  2. Were the roots of the first premolars an insurmountable obstacle to the alignment of the canines?
  3. Would further orthodontic treatment cause further resorption of the roots of the incisors, even though the incisors now required very limited movement, in the form of mild lingual tipping?
  4. Should I refuse further orthodontic treatment, remove the appliances and refer the patient for extraction of the canines and the four severely resorbed maxillary incisors and prosthodontic rehabilitation?

These difficult questions were discussed with the parents and the following answers were offered:

  1. The attempt should be made to bring the canines into alignment and de-rotated and it was expected that this could be done with a fair prognosis.
  2. The roots of the first premolars should be distally uprighted – over-uprighted – and deliberately rotated to an excessive degree in a mesio-palatal rotation. This would eliminate the obstacle that the curled palatal roots of these teeth presented for canine alignment. Following the canine alignment, the premolars would be returned to an upright position and at least partially de-rotated, as far as the canine root will allow2.
  3. Given that the canines were now distanced from the incisors, the principal etiologic factor for resorption had been eliminated and the resorption should stop.3 Furthermore, in the light of the fact that only a mild tipping movement of the incisors was contemplated, the risk of additional root resorption from this cause was considered to be minimal.
  4. Since alternative treatment modalities would require the extraction of the 6 anterior teeth, it was considered that there was little to be lost, even if the prognosis for the outcome of the orthodontic treatment was poor. The clinical situation in December 2006 could not be left untreated. The teeth were very prominent and of poor appearance. There was occlusion on the second molars alone, which meant that function was markedly compromised. Treatment could clearly not be refused.

It was pointed out to the parents that the accumulated prognoses of the several areas of contention could result in the achievement of a less-than-ideal outcome and this was accepted by them before treatment was resumed.


The treatment per se was unremarkable, except that the patient’s oral hygiene left much to be desired. The brackets on the first premolars were removed and Tip-Edge brackets bonded in their place, to provide the capability for distal root uprighting without producing an unwanted and potentially damaging intrusive effect on the lateral incisors. Elastic ties between the palatal canines and the passive heavy labial base arch were used to direct the canines to their places in the arch.


Fig. 5. a Tip Edge bracket has been substituted on the first premolar and a side-winder uprighting spring is in place. The funnel-shaped slot of this bracket and its wide tolerance allow the uprighting movement to occur on a rigid, passive, rectangular, base arch, without a reactive effect on the resorbed incisors.

Fig. 6. in March 2007, after premolar uprighting was achieved, new eyelets have been bonded to the anatomic labial surface of the canines and elastic ties to the labial arch are rotating the teeth while drawing them horizontally and labially towards the heavy, passive, base arch.

At about half way, the left canine became buried in the palatal mucosa of the lateral alveolar ridge and caused an acute periodontal condition, which was relieved by releasing the elastic tie, irrigating and placing an auxiliary labial arch (see October 2011 bulletin on this website) to extrude the tooth more vertically and improving the oral hygiene. At the next visit, lateral traction was re-instated, without further problems.


Figs. 7, 8. Because of the height differential between the canines and the occlusal plane, the labial movement has caused the canines to become buried in the palatal side of the tissues covering the vertical ridges, in July 2007. Labial traction was temporarily halted while an auxiliary arch was used to vertically erupt the teeth.

Fig. 9. In the final stages of labial traction and rotation in June 2008, a NiTi auxiliary wire is threaded through the vertically directed eyelets to complete the leveling and alignment stage.

The initial tipping of the canines to the labial archwire was completed and a Begg-type torqueing auxiliary was placed over the heavy main arch to torque the roots of the canines labially. The reader should note that the use of rectangular archwire to achieve this would have generated reverse torque on the adjacent severely compromised anchor teeth – the lateral incisors – which would have had a negative effect on their prognosis.


Fig. 10. a periapical view of the left canine region in January of 2009 shows the first premolar fully corrected, the canine fully aligned and the lateral incisor with the same severity of resorption as 2 years earlier. The bone height is relatively poor, but it should be remembered that new bone is not visible on radiographs until it become more matured.


Figs. 11 the clinical intra-oral views of the finished case at the time of appliance removal and splint placement.

Three months of orthodontic alignment of the teeth in the mandibular arch and the establishment of normal occlusion then completed the active treatment. Fixed bonded lingual twistflex retainers were placed on the anterior teeth in both jaws.


Fig. 12. The panoramic view at treatment completion.

p.a. Dec. 2009 -1_1p.a._Dec._2009__2p.a._Dec._2009__3

Fig. 13. The periapical views at treatment completion.

At the 1 year recall visit, the final dental alignment remained satisfactory, with the exception of the left maxillary canine, which had undergone some palatal positional relapse. The radiographs, taken 1 year after completion of all treatment, showed little or no alteration of root length compared with those taken in November 2006, at the start of this final phase of treatment.

Discussion and “post mortem”:

Accurate positional diagnosis of the initial location of the canines had been absent from the initial diagnosis of the case.4 Without question today, most of us would probably have performed a CBCT to achieve this5, 6. However, the CBCT was only performed post factum, after the surgical exposure had been carried out. Nevertheless, a careful study of the initial plane radiographs showed that they possessed enough information to know that the canine crowns were extremely high in the alveolus, above or at the level of the root apices of the incisors. The panoramic view taken in December 2003 shows the crown tip of the canine in close relation with the apices of the lateral incisors, but still distal to the central incisors. The cephalogram taken on the same day clearly outlines the canine crowns and shows them to be anteriorly situated , in a line just palatal to the central incisor roots. However, they are labial to the more lingually placed root apices of the lateral incisors.

A palatal approach was taken for the surgery and attachments were placed on that side of the canine crowns. The initially-applied direction of traction was vertically downwards which, for any crown that was in close association above or on the same level as the root apices, would mean that the crowns would be brought into contact with the root apices. This would result in slow progress in the resolution of the impacted canines, possible arrest of further movement and the strong likelihood of incisor root resorption, as seen in this case, where the canines were labial!

It is pertinent to note that the degree of root resorption evident in this case seems to have been more exaggerated than is to be expected in most similar circumstances. Moreover, it has been reported that examples of extreme resorption seem to occur almost exclusively among female patients.3

In this clinical scenario there are only two valid courses of action. If the canine is above the level of the root apices, they may be drawn posteriorly and horizontally towards the lingual cleats on the molar bands, before moving them vertically downward, to erupt them in the palate. The alternative is to draw them labially, but from a point of force application which is high in the labial sulcus, i.e. aimed at drawing them horizontally outwards, before bringing them vertically downwards on the labial side of the ridge. In either case, well-directed light force of a good range will deliver rapid results.

The initial position of the canines was in line with the lateral incisors yet, following the traction and eventual eruption into the mouth, the canines were situated very close to the midline raphe on each side. This is generally due to the use of a ballista arm or auxiliary labial arch loop (see October 2011 bulletin on this website) which is too long. As the result, the impacted tooth is drawn vertically downwards, but with a medial (towards the midline) component. It is important to use an arm or loop whose terminal helix is opposite the exit of the steel ligature through the flap – or shorter. By drawing the canines so far medially that they exit the tissues close to the midline raphe, means that they are also high in the palate and need to be drawn vertically downwards as they move buccally, for them to be accommodated in the dental arch. As such, they may become buried in the palatal soft tissue covering the vertical alveolar process and may produce an acute periodontal inflammation, which is very painful. This is what occurred in this case.

The grotesquely curved palatal root of the two premolars is not as infrequent a finding as might be thought. It is often missed on radiographs because of superimposition of the buccal root, but more often because the film is not examined carefully enough. Furthermore, when space opening is performed, the effect of a buccal coil spring on the archwire is to cause the premolar to tip distally and to rotate disto-palatally, which brings the palatal root more mesially. In the same way as preparation for an implant for prosthodontic reasons requires creation of space between the roots of the adjacent teeth, so too is this necessary when preparing for impacted canine alignment.


Teeth with short roots have a disadvantaged life expectancy….. and many would add that that is the understatement of the week! Nevertheless, this does not mean that they will fall out tomorrow and neither does it mean that they should be replaced by implants as soon as possible. One of the most important aims of retaining these teeth in the young patient is to preserve alveolar bone for future implants and to guide the patient through his/her growth period without the need for the fabrication of extraneous prosthetic temporary replacements, which are very unsatisfactory from almost every point of view. While this principle has been uppermost in my mind over the years, I have been in practice long enough to see that in most cases these teeth will last well into adult life. There are undoubtedly exceptions but, even with very short roots and once a few post-treatment months have passed for mature bone to form around them, they are rarely overly mobile and they seem to remain in place most satisfactorily – against all earlier dire predictions and warnings.

With the possible exception of the maxillary left canine, appearance in this case is excellent and there would seem to be no urgency to offer any form of prosthodontic alternatives to the patient for several years to come. At present, more than 2 years out of treatment and according to specific reports from her general dentist, there are no deleterious changes in the situation – although I have failed in my efforts to bring her in for routine follow-up since the treatment was completed.


  1. Jacoby H. The ballista spring system for impacted teeth. American Journal of Orthodontics 1979; 75: 143–51
  2. Becker A. The Orthodontic Treatment of Impacted Teeth 3rd ed. Oxford, John Wiley Blackwell, 2012 page 111, in press.
  3. Becker A, Chaushu S. Long-term follow-up of severely resorbed maxillary incisors following resolution of etiologically-associated canine impaction. American Journal of Orthodontics and Dentofacial Orthopedics 2005, 127: 650-654
  4. Becker A, Chaushu G, Chaushu A. An analysis of failure in the treatment of impacted maxillary canines. American Journal of Orthodontics and Dentofacial Orthopedics 2010;137:743-54.
  5. Chaushu S, Chaushu G, Becker A. The role of digital volume tomography in the imaging of impacted teeth. World Journal of Orthodontics, 2004; 5:120-13
  6. Becker A, Chaushu S, Casap-Caspi N. CBCT and the Orthosurgical Management of Impacted Teeth. Journal of the American Dental Association 2010;141(10 suppl):14S-18S