• allergy to formaldehyde;
  • anaphylactic shock;
  • angioedema;
  • anti-formaldehyde IgE;
  • endodontic treatment;
  • formaldehyde;
  • root canal sealant;
  • urticaria

The authors report seven cases of allergic reactions, four cases of anaphylactic shock and three of generalized urticaria, to formaldehyde contained in root canal sealant after endodontic treatment.

The clinical presentation, skin tests, high levels of anti-formaldehyde immunoglobulin E (IgE), as well as the study of the previous cases reported in the literature, suggest allergic IgE mediated mechanisms. These very infrequent but potentially severe reactions in endodontic therapy focus attention on the different manifestations related to formaldehyde, the involved mechanisms, the diagnostic procedure and the prevention possibilities in dentistry.

Formaldehyde (formalin, paraformaldehyde, trioxymethylene) is widely used in industry, cosmetics, disinfectants, medications and root canal sealants (1, 2).

The pathological reactions related to formaldehyde such as nasal, laryngeal and bronchopulmonary lesions appearing upon inhalation, gastrointestinal lesions appearing upon ingestion, and cutaneous necrosis or contact dermatitis, may be caused by simple irritant mechanisms (1–3). In addition, hypersensitivity or allergic reactions such as rhinitis, asthma, generalized urticaria, angioedema and anaphylactic shock, have been described (4–27).

Despite the high frequency of root canal treatments with sealant containing formaldehyde, very few cases of well documented allergy to root canal sealant have been reported in the literature. These sealants are a complex mixture of potentially irritating and/or sensitizing substances, such as metals, eugenol, formaldehyde, menthol, phenol, etc (5, 6).

Formaldehyde release from root canal sealant has been demonstrated in vitro and in vivo, and as a hapten may induce anaphylactic reactions after reacting with other proteins to become a complete allergen. (12, 18, 28–32).

The authors report seven cases of allergic reactions, four cases of anaphylactic shock and three of generalized urticaria, to formaldehyde contained in root canal sealant after endodontic treatment.

The clinical presentation, skin tests, high levels of anti-formaldehyde immunoglobulin E (IgE), as well as the study of the previous cases reported in the literature, suggest allergic IgE mediated mechanisms. These very infrequent but potentially severe reactions in endodontic therapy focus attention on the different manifestations related to formaldehyde, the involved mechanisms, the diagnostic procedure and the prevention possibilities in dentistry.

Case 1: A 41-year-old nonatopic man presented with a periapical granuloma of tooth 24 requiring endodontic treatment. Several minutes after complete dental treatment the patient complained of a sensation of warmth, generalized pruritus and respiratory difficulty. Thirty minutes later he developed anaphylactic shock with a drop in systolic blood pressure to 50 mmHg and lost consciousness. The outcome was favorable with an emergency resuscitation.

The root canal sealant used was Spad® (Quetigny, France), a mixture of powder and liquids with the following composition: phenylmercury borate, calcium hydroxide, hydrocortisone acetate, trioxymethylene, titanium oxide, barium sulfate, zinc oxide (powder), glycerin, resorcinol, hydrochloric acid (liquid 1) and glycerin and formaldehyde 87% (liquid 2).

Skin prick tests to common aeroallergens and latex performed after the accident were negative. The skin prick test to liquid 1 was negative and to liquid 2 was mildly positive, with a 3 mm diameter wheal vs 4 mm to codeine as a positive control. The same tests were negative in five control subjects. Patch tests, performed using finn chambers, to the standard battery of International Contact Dermatitis Research Group (ICDRG) allergens containing formaldehyde and to resorcinol (Trolab Allergenes, Reinbeck, Germany) were negative. Patch tests to the powder and two liquids (1% solution in Vaseline) induced a strongly positive delayed reaction with a confluent eczema beyond the test area. Anti-formaldehyde IgE were class 4 (25 kU/l: RAST CAP RIA, Pharmacia, Uppsala, Sweden).

Case 2: A 45-year-old woman underwent several root canal treatments with Spad® between 1987 and 1997 without any complications (tooth 16, 17 and 26). In 1997, a second endodontic treatment of tooth 26 with Spad® was followed by discomfort, anxiety, pruritus of the hands and pallor, which regressed after antihistamine treatment. The patient underwent treatment of tooth 16 for the second time in 1998 with Method Z® (Zizine France). Fifteen minutes later, she experienced tachycardia with extreme apprehension, pruritus and erythema of the hands and was unresponsive to antihistamines. This was followed by angioedema, dyspnoea and severe systolic hypotension (60 mmHg). The outcome was favorable with administration of adrenaline and systemic steroids.

The root canal sealant Method Z® contains: enoxolone, barium sulfate, excipient (powder), resorcinol, hydrochloric acid, excipient (liquid 1), and formaldehyde 35% and excipient (liquid 2).

Skin prick tests to common aeroallergens, latex and to liquid 1 and 2 at the concentration of 1% were negative. Patch tests with liquid 2 (1% solution) induced a local eczema after 48 h. RAST to formaldehyde was class 1 (0.41 KU/l; AlaSTAT DPC).

Case 3: A 43-year-old woman underwent an endodontic treatment of tooth 42 with Zial Z® (Zizine, France). Several similar procedures had been performed in the past. She experienced thoracic oppression and erythema developing within 24 h on two occasions. Two hours after the last treatment the patient developed generalized erythema, angioedema, vomiting, diarrhea, and hypotension and lost consciousness twice. After the first emergency resuscitation with adrenaline, she developed cardiac arrest whilst in hospital. The final outcome was favorable.

The root canal sealant Zial® contains: hydrocortisone, trioxymethylene, diiodothynol, E110, barium sulfate, zinc oxide, magnesium stearate (powder) and eugenol (liquid).

Skin prick tests and patch tests to liquid and powder at different concentrations were negative. Anti-formaldehyde IgE were class 3 (5.09 KU/l RAST CAP RIA Pharmacia Sweden).

Case 4: A 50-year-old man underwent several root canal treatments with Spad®. The treatment of tooth 45 in July 1996 was followed 2 h later by urticaria of the head and facial edema. In December 1996, 30 min after treatment of tooth 5, the patient developed pruritus, generalized urticaria, abdominal pain and discomfort. An allergy to the local anesthetic agent (lidocaine) was suspected but the skin tests and the challenge test were negative. In September 2002 the patient needed extraction of tooth 45 and endodontic treatment of tooth 44. This procedure was hemorrhagic and tooth 45 treated in 1996 was broken during the extraction. Fifteen minutes later, the patient presented with facial erythema, pruritus, generalized urticaria, abdominal pain, dyspnoea, discomfort with tachycardia and severe hypotension. The outcome was favorable with administration of adrenaline and steroids given several times in the emergency hospital. One month later, obturation of the pulp chamber of tooth 38 without apical treatment (pulpotomy) was well tolerated.

Skin prick tests to the powder and liquids of Spad® performed with 1% dilution and pure form as used in dentistry were negative. Anti-formaldehyde IgE were class 6 (>100 KU/l; Unicap Pharmacia Sweden).

Case 5: A 40-year-old woman, with a history of allergy to grass pollen and house dust mite, was treated for the second time for a granuloma of tooth 21 with Resoplast® (Pierre Roland, France) and Temp Bond®. Three hours after dental treatment the patient developed abdominal pain and pruritus of the scalp followed by urticaria of the face, neck, upper extremity and chest without hypotension. Outcome was favorable with symptomatic treatment in the emergency hospital.

Resoplast® has the following composition: deltahydrocortisone, bismuth nitrate (powder); benzalkonium chloride and formaldehyde (liquid 1) and sulfosalicylic acid and resorcinol (liquid 2). Temp Bond® (Kerr, Romulus, MI) contains zinc oxide and eugenol.

Skin prick tests to undiluted liquid 1 were positive giving a 7 mm diameter wheal and edema of the forearm. They were negative in four control subjects. Similar tests with liquid 2 and eugenol were negative. Anti-formaldehyde IgE were class 5 (98.5 KU/l; RAST AlaSTAT DPC).

Case 6: A 64-year-old woman who underwent endodontic treatment with Spad® on two previous occasions, had experienced moderate to severe local edema. On the third occasion, 4 h after the procedure, in addition to local edema, she developed nausea, vertigo and generalized urticaria, which persisted for 3 days despite antihistamine treatment.

Skin tests were not performed. Anti-formaldehyde IgE were class 5 (65 KU/l; RAST Alastat DPC).

Case 7: A 56-year-old man underwent root canal treatment of tooth 28 with Spad® in February 2000 after several previous uncomplicated endodontic procedures. Thirty minutes later he developed first a significant localized edema and than an edema of the whole face. Several hours later he presented with generalized urticaria that lasted 3 days.

Prick skin tests were weakly positive to formaldehyde. Anti-formaldehyde IgE were class 6 (>100 KU/l Unicap Pharmacia Sweden).

The adverse reactions to formaldehyde, such as respiratory (asthma and rhinitis) and cutaneous (contact dermatitis) reactions and anaphylactic shock in hemodialysis, are well documented. However, despite its widespread use, IgE dependant allergic reactions are rarely described (1–3, 14, 18, 27, 33, 34).

In odonto-stomatology the root canal sealants containing formaldehyde are still widely used. Different side effects related to endodontic treatment have been reported, such as infection, inflammation, necrosis, arthritis, paresthesia of the dental branch of the mandibular nerve, fungal caseous sinusitis etc. (5, 35–44). In dentistry, formaldehyde is used for its antibacterial activity, for devitalization of the tooth pulp and for its role in polycondensation of resorcine. (5, 11–14, 18). Release of formaldehyde from endodontic material has been known for a long time. Different in vitro and in vivo studies have shown a systemic diffusion of C14 labeled formaldehyde from endodontic material. The formaldehyde release may be enhanced by repetitive endodontic treatments, apicetomies, extraction of the treated tooth and dental overfilling with extrusion of sealant in the periapex or in the apical granuloma (5, 28–32, 35). Compared with the pulp chamber, which is relatively inert from an immunological point of view, the periapex constitute a network of vascular and nervous systems joining the tooth to the rest of the immune system (5, 6, 36, 38, 40, 43, 44).

Formaldehyde is a low molecular weight chemical which, acting as a hapten, may react with other molecules such as cutaneous proteins, serum proteins, proteins of the pulp chamber or of the periapex, or even with another component of the root canal sealant to become a complete allergen (3, 5, 12, 14, 15, 17, 18).

Despite the frequency of formaldehyde use and the number of endodontic treatments (453 000 in 1990 in Denmark for 5 million inhabitants) the allergic reactions to it in dentistry remain infrequent (5, 6, 18). They are probably underestimated in endodontic practice. Thirty-five cases of allergic reactions to formaldehyde (7–27) including seven personal observations have been described (Table 1). These allergic reactions can be of different severity ranging from local or focal reactions to life-threatening anaphylactic reactions: 15 cases of anaphylactic shock, 18 cases with urticaria and/or angioedema, nausea, dyspnoea, exanthema, pruritus and two cases with nonclearly defined symptoms. An additional case of formaldehyde related anaphylactic shock in a patient undergoing renal dialysis has been reported (33, 34). The symptoms could be of early onset, appearing within several minutes to 1 h after dental treatment (nine cases), or delayed, appearing from 2 to 24 h after the treatment (21 cases) (Table 1).

Table 1.  Case reports of immunoglobulin E dependant reactions to formaldehyde after endodontic treatment
Authors (references)Patients (gender/age)SymptomsTime of onset (h)RASTSkin tests
  1. AS, anaphylactic shock; AOE, angioedema; U, urticaria; NMD, not mentioned or not done.

Wedental (20)M/54AS2NMDNMD
Molina (15)M/35AS3NMD+
Ito (25)M/60AS0.7NMD+
Ebner (9)M/57AS1+
Ebner (9)F/33AS5–6+
Fehr (11)M/39AS0.5–2++
Gensau (12)F/43AS3++
Wantke (19)F/67AS10–12+
Sayama (26)F/39AS2NMD+
Modre (22)M/31AS5++
Kunisada (27)F/50AS8++
Case 1M/41AS0.5++
Case 2F/45AS0.25++
Case 3F/43AS2+
Case 4M/50AS0.25+
Bercher (21)M/NMDU3.5NMDNMD
Rousseau-Ducelle (16)M/30AOEFew hoursNMDNMD
Rousseau-Ducelle (16)F/37AOE+U3.5NMDNMD
Al Nashi (23)F/23AOE1NMD+
Burri (6)F/20AOE + UNMDNMD+
Drouet (8)F/NMDU4–6NMD+/−
Forman (24)M/57AOE4NMD
Ebner (9)M/57AOE+U10–12+
Fehr (11)F/40AOE + UNMDNMD+
Fehr (11)F/59AOENMDNMD+
Gensau (12)F/47AOE7+
Gensau (12)F/30U12+
El Sayed (10)F/37UFew hoursNMD+
Sporcic (17)F/52AOE + U9.5NMD+
Tas (18)M/53U0.25++
Case 5F/40U3++
Case 6F/64U4+NMD
Case 7M/56AOE+U0.5++
Ebner (9)M/NMDNot definedNMD+
Ebner (9)F/NMDNot definedNMD+
Total14 M15 AS9 < 1 H20 +19 +
21 F18 U/AOE21 > 2 H0 −11 −

The skin tests to formaldehyde are not standardized and may provoke even delayed severe systemic reactions (15, 34). Prick tests to 0.1 and 1% formaldehyde solution are often negative and are inconsistently positive to the pure solution as used in dentistry. Patch tests to the standard battery of ICDRG containing formaldehyde are very often negative. Those skin tests to the native solution used in dentistry, or to 1% formaldehyde solution, sometimes give a delayed positive reaction, but their clinical significance is difficult to establish in some cases (14, 18). Skin tests were positive in 19 cases and negative in 11 cases (Table 1).

The measurement of specific IgE to formaldehyde is an important diagnostic element and may suggest underlying allergic mechanisms. Positive RAST, often with higher class, was detected in all cases when it was analyzed (20 cases). However, in some cases specific IgE have been detected without associated clinical symptoms (2, 3, 44–46). This may raise the question of its real significance in view of ubiquitous exposure to formaldehyde, particularly by respiration. Formaldehyde in powder form or in aqueous solution may be more reactive than in gaseous form and thus may lead to sensitization in odonto-stomatology. This sensitization could result from domestic or occupational contact with formaldehyde (cosmetics, certain medicines, in dentists, anatomists, pathologists, etc) and especially after previous endodontic treatments (all seven in our case reports). Dental overfilling with extrusion of root canal sealant and also instrumental intervention (apicectomie, dental extraction of treated tooth, repeated treatments of the same tooth, etc) may promote diffusion of the soluble formaldehyde in the apical or periapical region which is inflammatory and hypervascularized in conditions such as apical granuloma (5, 28–30, 32, 35, 40–42). The more or less rapid diffusion of formaldehyde after the endodontic treatment and the necessity of binding with a protein to form an antigenic conjugate may explain the more or less rapid induction of anaphylactic shock in certain cases and the inconsistently positive skin tests (12, 18).

Despite the widespread domestic and occupational use of formaldehyde and the frequency of endodontic treatments with sealant containing formaldehyde, IgE dependant allergic reactions in dentistry appear to be rare but they could be potentially severe and life threatening. Their incidence is unknown and is perhaps underestimated in the literature.

Our cases, as well as those reported in the literature, suggest that in dentistry, in the case of an allergic reaction, it is important to consider formaldehyde contained in root canal sealant as an etiological agent, along with local anesthetic and latex.

There is a need to use biocompatible material which does not contain formaldehyde and which does not release any component in endodontic treatment.

Use of sodium hypochlorite 3% for disinfecting and obturation with gutta percha and/or cements or sealants without formaldehyde AH Plus (Detrey-Dentsply, Konstantz, Germany), Sealapex® (Kerr Romulus, MI, USA), Pulpispad (Spad, Quetigny, France) and avoiding apical extrusion of sealant can be proposed (5, 6, 18, 38, 39, 43, 44, 47, 48).


  1. Top of page
  2. References
  • 1
    Foussereau J. Guide de dermato-allergologie professionnelle. Paris: Masson, 1991.
  • 2
    Leroyer CH, Dewitte JD. Asthme au formaldehyde. In: BessotJC, PauliG, editors. L'asthme professionnel.Paris: Margaux Orange, 1999: 353363.
  • 3
    Smedley J Editorial. Is formaldehyde an important cause of allergic respiratory disease? Clin Exp Allergy 1996;26: 247249.
  • 4
    Braun JJ, Zana H, Bessot JC, De Blay F, Pauli G. Choc anaphylactique par allergie au formol d'une pâte canalaire lors d'un traitement endodontique. Revue française d'Allergologie 1998;38: 705708.
  • 5
    Braun JJ, Valfrey J, Scherer Ph, Zana H, Ha Y, Pauli G. Allergie IgE dépendante au formol de pâte canalaire lors du traitement endodontique. Rev Stomatol Chir Maxillofac 2000;101: 169174.
  • 6
    Hakel Y, Braun JJ, Zana H, Boukari A, De Blay F, Pauli G. Anaphylactic shock during endodontic treatment due to allergie to formaldehyde in a root canal sealant. J Endodontics 2000;26: 529531.
  • 7
    Burri C, Wüthrich B. Quincke-Ödem mit Urtikaria nach Zahnwurzelbehandlung mit einem paraformaldehyd-haltigen Dentalantiseptikum bei Spättyp-Sensibilisierung auf Paraformaldehyd. Allergologie 1985;8: 264268.
  • 8
    Candura F. Formaldehyde-induced anaphylaxis after dental treatment. Letter to the Editor. Contact Dermatitis 1991;25: 335.
  • 9
    Drouet M, Le Selin J, Bonneau JC, Sabbah A. Allergie à la pâte canalaire. Allergie et immunologie 1986;18: 4143.
  • 10
    Ebner H, Kraft D. Formaldehyde induced anaphylaxis after dental treatment. Contact Dermatitis 1991;24: 307309.
  • 11
    El Sayed F, Seite-Bellezza D, Sans B, Bayle-Lebey P, Marguery MC, Bazex J. Contact urticaria from formaldehyde in a root canal dental paste. Contact Dermatitis 1995;33: 353.
  • 12
    Fehr B, Huwyler T, Wüthrich B. Formaldehyde and paraformaldehyde allergy. Allergic reactions to formaldehyde and paraformaldehyde. Schweiz Monatssch Zahnmed 1992;102: 6467.
  • 13
    Gensau A, Pirkhammer D, Aberer W. Anaphylaxie durch parafomaldehydehaltige Dentalmaterialen. Allergologie 1994;9: 439441.
  • 14
    Kränke B, Aberer W. Formaldehyd und Paraformaldehyd in der Zahnmedizin als Ursache Schwerer anaphylacto Reaktionen. Allergologie 1997;5: 246251.
  • 15
    Molina C, Passemard N, Godefroid JM. Allergie au formol et odontostomatologie. Revue Française d'Allergologie 1971;11: 1118.
  • 16
    Rousseau-Decelle. Deux cas d'oedème de Quincke et d'urticaire généralisée consécutifs à l'emploi de trioxymethylene. Rev Stomatol 1936;38: 569.
  • 17
    Sporcic Z, Paranos S. Allergy to a tooth devitalizing material. Allergy 2001;56: 249.
  • 18
    Tas E, Pletscher M, Bircher AJ. IgE-mediated urticaria from formaldehyde in a dental root canal compound. J Invest Allergol Clin Immunol 2002;12: 130133.
  • 19
    Wantke F, Hemmer W, Halgmüller T, Götz M, Jarisch R. Anaphylaxis after dental treatment with a formaldehyde-containing tooth-filling material. Allergy 1995;50: 274276.
  • 20
    Wedendal PA. Allergic shock following root canal treatment with tricresol-formalin. Svensk Tändläk-T 1945;47: 319321.
  • 21
    Bercher J. Un cas d'urticaire récidivante après l'emploi de pâte rose. Rev. Stomatol 1936;38: 577580.
  • 22
    Modre B, Kränke B. Anaphylactic reaction to formaldehyde. Allergy 2001;56: 263264.
  • 23
    Al Nashi YG, Al-Rubayi A. A case of sensitivity to tricresol formalin. Br Dent J 1977;142: 52.
  • 24
    Forman Gh, Ord RA. Allergic endodontic angioedema in response to périapicale endomethasone. Br Dent J 1986;160: 348350.
  • 25
    Ito M, Sai M, Handa Y. Allergic reaction to formaldehyde contained in formocresol. J Dent Med (in Jap) 1988;28: 897904.
  • 26
    Sayama S, Tanabe H, Kizaki J. A case of anaphylactic shock caused by dental paste for root canal. Jpn J Clin Dematol (in Jap) 1996;50: 10671069.
  • 27
    Kunisada M, Adachi A, Asano H, Horikawa T. Anaphylaxis du to formaldehyde released from root canal disinfectant. Contact dermatitis 2002;47: 215218.
  • 28
    Araki K, Isaka H, Ishii T, Suda H. Excretion of 14C-formaldehyde distributed systemically through root canal following pulpectomy. Endodontics and Dental Traumatology 1993;9: 196199.
  • 29
    Block RM, Lewis RD, Hirsch J, Coffey J, Langeland K. Systemic distribution of 14C-labeled paraformaldehyde incorporated with formocresol following pulpotomies in dogs. Journal of Endontics 1983;9: 176189.
  • 30
    Hata G, Nishikawa J, Kawazoc S, Toda T. Systemic distribution of 14C-labeled formaldehyde applied in the root canal following pulpectomy. J Endodontics 1989;15: 539543.
  • 31
    Koch MJ, Wünstel E, Stein G. Formaldehyde release from ground root canal sealer in vitro. J Endodontics 2001;27: 396397.
  • 32
    Myers DR, Shoaf HK, Dirksen TR, Pashley DH, Whiford GM. Distribution of 14C-formaldehyde after pulpotomy with formolcresol. J Am Dent Assoc 1978;96: 805813.
  • 33
    Bousquet J, Rivory JP, Maurice F, Skassabrociek W, Larrson P, Johansson SGO et al. Allergy in chronic haemodialysis. A double blind intravenous challenge with formaldehyde. Clin Allergy 1987;17: 499506.
  • 34
    Maurice F, Rivory JP, Larsson PH, Johansson SGO, Bousquet J. Anaphylactic shock caused by formaldehyde in a patient undergoing long-term haemodialysis. J Allergy Clin Immunol 1986;77: 594597.
  • 35
    Bergenholtz G, Lekholm U, Milthon R, Engstrom B. Influence of apical overinstrumentation and overfilling on re-treated root canals. J Endodon 1979;5: 301310.
  • 36
    Bogaerts P, Simon JHS. Absence de guérison après traitement endodontique adéquat. Rev Belge Méd Dent 1992;4: 101115.
  • 37
    Eriksen HM, Bjertnes E, Orstavic D. Prevalence and quality of endodontic treatment in an urban adult population in Norway. Endod Dent Traumatol 1988;4: 122126.
  • 38
    Lin L, Skribner JE, Gaengler P. Factors associated with endodontic treatment failures. J Endodon 1992;12: 625627.
  • 39
    Mallouf EM, Gutmann JL. Biological perspectives on the non-surgical endodontic management of periradicular pathosis. Int Endod J 1994;27: 154162.
  • 40
    Odesjo B, Hellden L, Salonen L, Langeland K. Prevalence of previous endodontic treatment, technical standard and occurence of periapical lesions in a randomly selected adult, general population. Endod Dent Traumatol 1990;6: 265272.
  • 41
    Riccucci D. Apical limit of root canal instrumentation and obturation. Part 1. Literature review. Int Endod J 1998;31: 384393.
  • 42
    Riccucci D, Langeland K. Apical limit of root canal instrumentation and obturation. Part 2. A histological study. Int Endod J 1998;31: 394409.
  • 43
    Sjogren U, Hagglund B, Sundqvist G, Wing K. Factors affecting the long-term results of endodontic treatment. J Endodon 1990;10: 498504.
  • 44
    Smith CS, Setchell DJ, Harty FJ. Factors influencing the success of conventional root canal therapy-a five year retrospective study. Int Endod J 1993;26: 321333.
  • 45
    Patterson R, Pateras U, Grammer LC, Harris KE. Human antibodies against formaldehyde-human conjugates or human serum albumin in individuals exposed to formaldehyde. Int Arch Allergy Appl Immunol 1986;79: 5361.
  • 46
    Wantke F, Focke M, Hemmer W, Tscabitscher M, Gann M, Tappler P et al. Formaldehyde and phenol exposure during an anatomy dissection course: a possible source of IgE-mediated sensitization. Allergy 1996;57: 837841.
  • 47
    Ha Y, Wittenmeyer W, Bateman G, Bentaleb A, Allemann C. A new method for the quantitative analysis of endodontic microleakage. J Endodontics 1999;25: 172177.
  • 48
    Watts A, Paterson RC. ‘‘Usage’’ testing of root-canal sealing materials. A critical review. J Dent 1992;20: 259265.