A. Brynjulfsen, I. Fristad, T. Grevstad & I. Hals-Kvinnsland
Department of Odontology – Endodontics, School of Dentistry, University of Bergen, Bergen, Norway.
Department of Anatomy and Cell Biology, Medical Faculty, University of Bergen, Bergen, Norway.

Introduction.
Incompletely fractured teeth associated with pain, have been referred to as a diagnostic challenge (Cameron 1964). Various terms have been used to describe this type of tooth fracture, such as ‘green-stick fractures’ (Sutton 1962), ‘split root syndrome’ (Silvestri 1976) and ‘cracked tooth syndrome’ (Cameron 1964). The incompletely fractured tooth has been defined as ‘a tooth with a demonstrable fracture, but no visible separation of the segments’ (Luebke 1984). These dentinal injuries may be asymptomatic, or associated with symptoms and even orofacial pain (Cameron 1964, Homewood 1998, Ailor 2000).
Scarce information is available about the incidence of incompletely fractured teeth. In a clinical study of 1194 molars and premolars (Motsch 1992), incomplete fractures were found in about 3% of the teeth after removal of amalgam fillings. Furthermore, pain was rarely observed with this type of tooth injury, indicating that incompletely fractured teeth mainly exist without symptoms.
Incompletely fractured teeth may, in typical cases, be fairly simple to identify by common test procedures, such as selective bite-tests, use of staining solutions and transillumination. In other cases, however, they may be very difficult to localize because pain symptoms mimic other possible diagnoses like symptoms related to sinusitis, temporomandibular joint disorders, headaches, ear pain, or atypical orofacial pain (Snyder 1976). Traditional diagnostic tools may be less helpful in such cases, and have to be supplemented.
The purpose of this study was therefore to present findings from patients with diffusely localized orofacial pain, investigating the diagnostic value of different clinical findings, and analysing the distribution and pattern of pain experienced from incomplete dentinal fractures. Secondly, the effect of various treatment modalities were of interest.

Materials and methods.
Thirty-two patients referred to the teaching clinic at the Faculty of Dentistry, University of Bergen with diffuse orofacial pain were finally diagnosed with 46 incompletely fractured teeth. Those included in the study represent a selected group of patients with one or more incompletely fractured teeth and associated diffuse orofacial pain.
The 32 patients were referred from general dental practitioners, oral surgeons, periodontists, general medical doctors and medical specialists. More than 50% of them had been suffering diffuse orofacial pain for more than 1 year, and had earlier consulted both dental and medical expertise. Yet, no specific diagnosis had been made.
They were now examined according to standardized questions, radiographic evaluation and clinical procedures, including percussion test, selective bite-test, use of staining solutions and transillumination. None of these patients, however, experienced the typical pain pattern normally associated with incompletely fractured teeth, i.e. pain of sudden onset and short duration provoked by cold and chewing pressure. The information was recorded in a form used at the dental clinic, specially designed for pain evaluation.
The patients were then subjected to a thorough examination in search of suspected incomplete fractures. The usual testing procedures, performed initially, did not directly aid in the detection of a possible incompletely fractured tooth. Radiographs were of little value in detecting an incompletely fractured tooth, but were taken in order to rule out other possible diagnoses. In cases with unilateral diffuse pain spreading to both maxilla and mandible, anaesthesia tests were of great help in determining the correct jaw. Thereafter, fillings were removed routinely, starting with the most posterior tooth, allowing direct inspection of possible fracture lines. Removal of fillings was stopped when a fracture line was discovered. Removal of fillings located more mesially was continued if symptoms persisted after treatment. Methylene blue was, in many cases, used to stain a suspected crack. Microscopic evaluation and transillumination of the stained area provided even more information.
Once the fracture lines were revealed, further information was recorded; source of patient referral, affected jaw, affected tooth, extension of the fracture lines, type of restoration present, number of incompletely fractured teeth in each patient, duration of pain, and finally distribution and pattern of pain experienced by patients with one or more diagnosed incompletely fractured teeth in the same quadrant and by patients with affected teeth on both sides of the midline.

Pathways of referrals to the endodontists.
The referred patients had consulted their own general practitioner (GP) due to pain problems in the orofacial region, 26 patients more than twice. Nineteen patients (63%) were referred directly from the GP to the endodontist, four (9%) from oral surgeons, two from periodontists (6%) and seven (22%) from medical services. In the latter group, five patients had consulted an ear, nose and throat physician and two patients had consulted a neurologist. Three patients in this group were hospitalized. Still, the correct diagnosis was not made.

Incidence of crack lines.
Analysis of the 46 teeth showed that the majority of the incompletely fractured teeth occurred in the maxilla, evenly distributed between premolars and molars, whereas molars predominated in the mandible (Fig. 1).
Twenty-three patients (72%) were diagnosed with only one incompletely fractured tooth. Nine patients (28%) exhibited incomplete fractures in more than one tooth, ranging from two to six. In four of these patients fracture lines were located in the same quadrant, and in five patients they occurred on both sides of the midline. No patients were recorded with infracted teeth in both jaws on one side only.
The majority (89%) of the incompletely fractured teeth occurred in heavily restored teeth (Fig. 2). Three infractions (7%) were found in teeth with class I restorations, and two (4%) in teeth without restorations.

Figure 1. Distribution of infracted maxillary and mandibular teeth with symptoms. _ = maxilla, _ = mandible.
P1, first premolar;
P2 second premolar;
M1, first molar;
M2, second molar.

Extension of fracture lines.
All fracture lines extended in the mesio-distal direction, and in 10% of the cases they were running centrally in the tooth.
In maxillary teeth, 70% of the infractions were localized in the dentine supporting the buccal cusps, whilst 20% were on the lingual side. In the mandible, the situation was opposite, 20% buccally and 70% lingually.

Duration of pain prior to treatment.
Twenty-five patients had been suffering pain for more than 3 months, one for as much as 11 years (Fig. 3). No connection was observed between duration of pain and the number of incompletely fractured teeth. Likewise, the duration of pain did not negatively effect the treatment outcome.

Distribution of pain.
In cases with incompletely fractured mandibular teeth, the pain was felt throughout the arch and in tissues including neck, ear, chewing muscles and TMJ on the same side. In addition, it was regularly projected to maxillary teeth and tissues. In most cases the patient felt that the pain was located anterior to the affected tooth.
When one or more incompletely fractured teeth were diagnosed only in the maxilla, pain projecting to the mandible was reported infrequently. The pain was usually located anterior to the affected tooth. Location to the ear-region was infrequent.
The longer the duration of pain before the diagnosis of an incompletely fractured tooth was established, the more diffuse was the distribution of pain. Often symptoms projected to both jaws.
In five patients, incomplete dentinal fractures occurred on both sides of the midline, in three of them in the maxilla. In these cases, pain was diffusely distributed throughout the mandibular and maxillary tissues including neck, ear, chewing muscles and TMJ on both sides. Headaches occurred in the group of patients with symptoms for more than 5 years prior to treatment.

Figure 2. Symptomatic incomplete dentinal fracture under an amalgam filling.
(a) An incomplete dentinal fracture (arrow) is seen after removal of an amalgam filling.
(b) Access preparation reveals a fracture line involving the pulp chamber.

Treatment.
Endodontic or restorative treatment relieved the symptoms in 90% of the patients during a two-year follow-up, whilst persisting symptoms in 10% were considered as part of an orofacial pain complex of unknown aetiology. The patients who had been hospitalized were all made symptom free.
If symptoms persisted after temporary treatment, including reduction of occlusal load and stabilization of the fracture lines, a diagnosis of irreversible pulpitis was made and root canal treatment was performed without delay. About half of the group was treated in this way, whilst the other half was treated with either cast restorations (cemented) or conventional fillings, always with cuspal coverage (Fig. 4).

Figure 3. of pain prior to treatment. _ represents acute pain; represents pain of chronic character.



Figure 4. Types of treatment performed.

Discussion.
In our selected material, 28 out of 32 patients with incompletely fractured teeth and associated pain had suffered pain for more than 3 months, most of them for years. Despite numerous visits to different dental and medical specialists, the correct diagnosis was not made. This may indicate a lack of knowledge concerning this special problem, or it may be other contributing factors making the diagnosis difficult. The fact that some patients had multiple infractured teeth, ranging from two to six infractions, makes the pain pattern in this group confusing and the diagnostics difficult. If incompletely fractured teeth were present on both sides of the midline, and the patient presented with a history of numerous visits to dentists and/or medical doctors, the patient most often arrived at our clinic with the diagnosis ‘atypical pain’, and the real cause remained undetected. We advocate that the dentist should suspect an incompletely fractured tooth in all cases with obscure aetiology and mixed chronic symptoms.
The pain pattern in patients with one diagnosed incompletely fractured tooth was not distinguishable from patients with more infractions in one quadrant.

Incompletely fractured teeth related to type of restoration.
The majority of the infractions occurred in heavily restored teeth. Many studies (Mondelli et al. 1980, Abou-Rass 1983, Cavel et al. 1985), have shown how cavity design affects the strength of the tooth.
In teeth with extensive amalgam restorations, systematic removal of fillings proved to be the most efficient way of tracing incompletely fractured teeth in our material.
The most difficult diagnostic cases, were incomplete dentinal fractures in intact teeth and in teeth with class I restorations. Those patients had reported pain for a period of 2–10 years, and the pain was often projected to the ear region, radiating to both mandibular and maxillary regions. The affected teeth were, in these cases, premolars.
The best method for detecting infractions in intact teeth and in teeth with a small restoration, is transillumination by use of fibre-optics. The tooth should be properly cleaned before placing the light source directly against the tooth surface.

Treatment or observation?
Pain connected to incompletely fractured teeth is usually not acute, and does not require immediate treatment. The dentist should not feel compelled to treat all cases with symptoms. Misdiagnosis and mistreatment may easily follow. In this material we observed ‘unnecessarily’ root-filled and extracted teeth in search of the affected tooth, if there was any.
On the other hand, it is important to prevent the pain from becoming chronic. Sessle et al. (1986) described how neurones in the brain stem receiving input from skin, oral mucosa, neck muscles and TMJ become operative in longstanding pathophysiological situations, neurones which are normally ineffective in exiting the nociceptive transmission neurones. Thus, both peripheral and central sensitization are involved to varying degrees in different types of hyperalgesia (Treede & Magerl 1995).
In our material, we found that in cases with longstanding pain, the perception of pain changed and the pain projected more diffusely throughout orofacial tissues, thus making the diagnosis more difficult. This indicates the importance of a thorough evaluation in the search for a diagnosis at an early stage in cases with orofacial pain. The specially designed form for pain evaluation used at our clinic was of great help in determining whether a dental cause of pain was present or not.

Treatment considerations.
 The treatment principles used in this study were in line with recommendations given by Homewood (1998). In about 50% of the cases, the diagnosis of irreversible pulpitis was determined, and endodontic treatment was performed. All of the incompletely fractured teeth with infraction lines running centrally in dentine were in this group.
Further studies may be necessary in order to determine whether teeth with centrally running infractions and a history of pain should be more readily root filled to avoid recurrence of pain. Later follow-up studies from this patient group support such a view.
In conclusion, this study shows that diagnosis of incompletely fractured teeth with symptoms is time consuming and represents a diagnostic problem. With appropriate endodontic and/or restorative treatment, symptoms were relieved in the majority of cases.

References.

Abou-Rass M (1983) Crack lines: the precursor of tooth fractures - their diagnosis and treatment. Quintessence International 4, 437-47.
Ailor JE (2000) Managing incomplete tooth fractures. Journal of the American Dental Association 131, 1168-74.
Cameron CE (1964) Cracked tooth syndrome. Journal of the American Dental Association 68, 405-11.
Cavel WT, Kelsey WP, Blankenau RJ (1985) An in vitro study of cuspal fracture. Journal of Prosthetic Dentistry 53, 38- 42.
Homewood CI (1998) Cracked tooth syndrome - Incidence, clinical findings and treatment. Australian Dental Journal 43, 217-22.
Luebke RG (1984) Vertical crown-root fractures in posterior teeth. Dental Clinic of North America 28, 883-95.
Mondelli J, Steagall L, Ishikiriama A, de Lima Navarro MF, Soares FB (1980) Fracture strength of human teeth with cavity preparations. Journal of Prosthetic Dentistry 43, 419- 22.
Motsch A (1992) Pulpitische Symptome als Problem in der Praxis. Deutsche Zahnarztliche Zeitung 47, 78-83.
Sessle BJ, Hu JW, Amano N, Zhong G (1986) Convergence of cutaneous, tooth pulp, viseral, neck and muscle afferents onto nociceptive and non-nociceptive neurones in trigeminal subnucleus caudalis (medullary dorsal horn) and its implications for referred pain. Pain 27, 219-35.
Silvestri AR (1976) The undiagnosed split-root syndrome. Journal of American Dental Association 92, 930-5.
Snyder DE (1976) The cracked tooth syndrome and fractured posterior cusp. Oral Surgery. Oral Medicine Oral Pathology 41, 698-704.
Sutton PRN (1962) Greenstick fracture of the tooth crown. British Dental Journal 112, 362-6.
Treede RD, Magerl W (1995) Modern concepts of pain and hyperalgesia: Beyond the polymodal C-nociceptor. NIPS 10, 216-28.