REVIEW ARTICLE

Payman Hamadani, Nicholas Chandler

A modern look at the Cvek PULPOTOMY

KEY WORDS
Cvek pulpotomy, direct pulp cap, mineral trioxide aggregate (MTA)

ABSTRACT
The Cvek pulpotomy procedure has been in use for over 40 years since the technique was first
published in 1978. The original technique remains unchanged; however, the materials used have
evolved considerably. This article explores the developments in materials and the clinical proced- ure
since the late 1970s and discusses how they might influence outcome. Two cases are present- ed to
demonstrate common clinical scenarios that a general dental practitioner may encounter in everyday
clinical practice. The treatment performed with the available materials is critiqued.

INTRODUCTION outcomes might be predicted today.

The Cvek pulpotomy procedure is primarily

aimed at maintaining pulp vitality in teeth with
compli- cated crown fracture. It has particular
application in managing traumatic pulp
exposure in immature teeth where continued
root maturation is desired. It involves removing
the most inflamed portion of the exposed
dental pulp and placing a pulp cap followed by
a restoration. The outcome is highly predictable
and unrelated to the size of the exposure and
the period between the accident and the
treatment.
Cvek’s landmark paper in 19781 reports
60 complicated crown fractures in permanent
inci- sors in patients aged 7 to 16 years old.
Almost half of these teeth (28) had immature
roots. A success rate of 96% was reported after
an aver- age follow-up period of 31 months.
The treatment procedure and materials used
in the original study have changed
considerably.
The present article looks at the Cvek
pulpotomy procedure in light of the current
literature, to con- sider modern approaches to
this well-established and successful technique.
As a result of improve- ments, excellent

ENDO EPT 2019;13(4):287–293 1

Materials and methods of Cvek
The aim of the 1978 study by
Cvek was to clinically and
radiologically assess the
frequency of healing of
traumatic pulp exposure after
a partial pulpotomy, taking
into consideration the size of
the pulp expo- sure and the
interval between accident
and treat- ment with respect
to the stage of root
development. In Cvek’s
study1, 60 anterior teeth
comprising
51 maxillary and 9
mandibular incisors in
patients aged between 7
and 16 years were assessed.
As a result of the traumatic
injury, some teeth had no
abnormal mobility, while
others had increased
movement with/without
tenderness to percussion.
However, the exact nature
of the luxation injuries was
not specified, and only the
size of the pulp exposure
and the time interval from
accident to treatment were
reported. There was no
mention as to whether
there was previous caries or
his- torical trauma to the
injured teeth. Most of the
complicated fractures
involved either a transverse
or oblique fracture (56
teeth), and four teeth had
longitudinal crown-root
fractures with the apical
extent of the fracture not
specified. A bleeding or
proliferative pulpal
response was observed in all
cases. Electric sensibility
testing of the teeth was

2 ENDO EPT 2019;13(4):287–293

Hamadani and Chandler A modern look at the Cvek pulpotomy

carried out. A positive response was reported the approval from an ethics
for all teeth preoperatively, even though committee and would be limited to
bleeding of the pulp tissue was evident. only one operator to eliminate
The pulpotomy treatment was carried out
dur- ing two visits and by 11 dentists. The first
visit involved the use of a high-speed diamond
bur and sterile saline to remove the most
coronal aspect of the pulp tissue, leaving a 2
mm deep cavity. Bleeding was controlled with
saline followed by the placement of Calasept
(Directa AB, Upplands Väsby, Sweden), a non-
setting calcium hydrox- ide (CH) dressing,
followed by a zinc oxide and eugenol (ZOE) as
a provisional filling material. The sizes of the
exposures were measured but the method used
was not described; wound sizes were between
0.5 to 4 mm.
At 3-week, 3- and 6-month reviews, radio-
graphs were used to assess the presence of a
hard tissue barrier, and a stage-two procedure
was car- ried out. It can only be assumed that
patients did not have the coronal tooth structure
aesthetically restored until this stage-two
intervention, which in some cases was months
apart. With a sharp probe, the calcified barrier
was assessed for conti- nuity. The hard tissue
was then covered with Dycal (Dentsply DeTrey,
Konstanz, Germany), a hard set- ting CH
material. Subsequently, a composite resin
restoration was placed; this was conducted
with either a two-paste chemically-cured
material or an early light-cured resin. The use
of a liner or base material was not mentioned.
An average follow-up period of 31 months and a
96% success rate were reported. The time
elapsed after injury (up to an average of 8
days) did not influence the outcome. A longer
period from injury to treatment involved just
three teeth as most patients sought treatment
within the first week. The partial pulpotomy
was successful in cases of closed (less than 0.5
mm) and open (more than 0.5 mm) apices. The
technique for measuring the apical opening was
not detailed.

Modern TECHNIQUE and materials

A study of a similar nature would now require
inter-operator variability. The initial assessment of the
injury would have entailed a detailed preopera- tive
medical history, dental history, nature of injury (blunt- or
sharp-force trauma) and an assessment of the pulp and
periapical health status of teeth, and considering the
possibilty of more extensive alveolar and soft tissue
injuries.
The pulp exposure would have been digitally
photographed, to allow measuring the size of the
exposure on a computer screen and by using imaging
software. A baseline record would have involved pulp
tests, primarily with CO2 snow; teeth would have been
assessed for luxation injuries and a peri- odontal depth
probing would have been performed. Furthermore, a
minimum of two periapical radio- graphs with a cone-
shift technique would have been indicated to check for
additional injuries, such as root fractures and possible
tooth luxation. Concomitant injuries may affect prognosis,
while teeth with ‘open’ apices have a more extensive
blood supply and are less susceptible to the loss of vitality
following a luxa- tion injury.
Currently, treatment can be performed in a
single visit. While the principles of the pulpotomy
procedure are basically the same, pulpal haemor- rhage
is now managed with a cotton pellet soaked in sodium
hypochlorite (NaOCl). Many bioactive materials are now
available that have been shown to improve long-term
maintenance of the pulp vitality in direct pulp capping
(DPC) experiments. These materials are more stable
over time and are more effective sealants than those
previously used. Haemostasis should be achieved with 3%
to 6% NaOCl after 5 to 10 minutes of direct contact with
the bleeding pulp tissue. If bleeding continues and
exceeds this time duration, then the pulp is likely to be
irreversibly inflamed and a full pulpot- omy or
pulpectomy may be recommended. NaOCl in
concentrations from 1.5 to 5.25% is currently regarded as
a safe, effective and as an inexpensive haemostatic
solution for DPC and partial and com- plete pulpotomy
procedures2,3. The antimicrobial properties of NaOCl
solution provide haemosta- sis and disinfection of the
dentine-pulp interface, chemical amputation of the
blood clot and fibrin, biofilm removal as well as removal
of damaged
cells at the mechanical exposure site.
 Hamadani and Chandler A modern look at the Cvek pulpotomy

An improved understanding of the dentine- pH of acids, has antibacterial properties and promotes defence
pulp complex allows clinicians to harness its mechanisms and repair. Unfortunately,
poten- tial and maintain pulp vitality and
continued root maturation. The transforming
growth factor-beta (TGF-) family of growth
factors is responsible for the upregulation of
surviving odontoblasts and recruitment of
fibroblasts to secrete reparative dentine at the
site of injury4. These growth fac- tors can also
be secreted by odontoblasts and pro- moted by
bioactive materials such as CH, or silicate
cements, for example, mineral trioxide aggregate
(MTA) (Dentsply Tulsa Dental, Tulsa, OK, USA),
or Biodentine (Septodont, Saint-Maur-des-Fossés,
France). The formation of dentine is not
possible without odontoblasts. However, a
hard tissue bar- rier is formed after pulpotomy.
This hard tissue barrier or ‘bridge’ is secreted
by fibroblasts that are recruited to the site of
injury. The mineralised tissue is heterogenous,
amorphous and atubular in nature and so,
histologically, not dentine5.
The size of the pulp exposure has no
significant bearing on the final outcome of DPC
or the Cvek pulpotomy procedure, but the size
of the pulp exposure is often difficult to
estimate clinically6. This may influence the
decision-making process and result in more
extensive treatments, such as pulpectomy and
root canal treatment. However, the remaining
tooth structure and the injury are important
factors to take into account in the over- all
treatment plan context.

Materials and methods for vital PULP

therapy
One major determining factor in the healing
pro- cess of pulp tissue is the presence or
absence of microorganisms. In the absence of
microorgan- isms, exposed pulps in rats were
successfully sealed with mineralised tissue even
without the place- ment of a medication or a
restoration7.
Aqueous CH has been considered as the
standard material for vital pulp therapy for
many years8-11. Its high alkaline pH, which
stimu- lates fibroblasts and neutralises the low
CH dissolves over time. The constitu- ents comprise a hydraulic
Tunnel defects have been calcium silicate powder containing oxide
demonstrated in the resultant compounds, including those of calcium, iron,
hard tissue barriers. CH is silicon, sodium, potassium, mag- nesium and
also absorbable and aluminium15. The favourable phys- icochemical
dimensionally unsta- ble12. characteristics stimulate the tissue reparative
The slow degradation of CH processes by recruiting and activating hard
may lead to microleakage, tissue-forming cells. The by-products formed
allowing microorganisms to during hydration of mixed MTA include CH and
spread through the defects. calcium silicate hydrate, which sustain an
Pulp repair using aqueous alka- line pH environment for prolonged
CH show increasing failure periods. Blood does not affect the setting of
rates over time, as the ma- MTA16. There is an excellent marginal
terial is resorbed and adaptability of MTA to dentine, with
deteriorates. components penetrating the tubules to give
Non-aqueous hard- adhesion, which is comparable to a glass iono-
setting CH materials (cal- mer cement (GIC). MTA promotes a biocompat-
cium salicylate ester ible, non-cytotoxic, antimicrobial environment
cements) are less suitable for and favourable surface morphology for bridge
pulp capping due to their forma- tion. A significant disadvantage of MTA,
limited release of hydroxyl of spe- cial relevance to the Cvek pulpotomy
ions. Their pH is usually in anterior
lower and their antimicro-
bial effect weaker.
Materials such as Dycal
(Dent- sply DeTrey) and Life
(Kerr Corp, Romulus, MI,
USA) suffer long-term
disintegration, and may fail
to support an overlying
permanent restoration13.
Light-cured liners and base
cements with CH addi- tives
have been developed (e.g
TheraCal LC [TLC, Bisco,
Schaumburg, IL, USA] and
Ultrablend Plus [Ultradent,
South Jordan, UT, USA]).
Their applica- tion is simpler,
but despite their resin
content, light- cured
materials have poor
mechanical strength and
their pH is low. Also, only a
small amount of cal- cium
ions is released, and this
group of materials is
considered cytotoxic . 14

Most of our current

understanding of vital pulp
therapy is based on the first
MTA material, Pro- Root
MTA (Dentsply Tulsa Dental).
Hamadani and Chandler A modern look at the Cvek pulpotomy

teeth, is coronal tooth discolouration. Its ture, and current etching systems
constitu- ents, which are metals such as produce excellent bond strengths to
bismuth oxide or iron, may oxidise and enamel, dentine and cured DPC
promote this adverse effect. Other calcium
silicate-based cements contain zir- conia or
tantalum oxide as radiopacifiers and these are
more colour-stable; hence less likely to cause
tooth discolouration.
Many newer calcium silicate cements show
physicochemical and bioinductive properties
com- parable to MTA17-19. The most popular
product among nearly 40 available on the
market is Bio- dentine, which demonstrates
strong bioactive and antibacterial properties20.
ZOE, used by Cvek in 1978, has traditionally
been used as a base under restorations or as a
pro- visional material21. ZOE has strong
antibacterial activity against Streptococcus
mutans and other microorganisms within
infected dentinal tubules22. Although eugenol
has been shown to have an inhib- itory effect on
the polymerisation of the composite resin, it can
still be considered as a suitable base, but a
bonding agent is essential to avoid polymerisation
shrinkage-induced detachment21.
GICs were in their infancy in the mid-1970s
and at that time were supplied as powder and
liquid formulations for spatulation. Today, an
interposed liner of a capsulated GIC
formulation is used to protect the capping
material, as well as to bond to the MTA and
the composite resin. Dentine adhe- sives and
composites are not biocompatible and should
not be used for pulpotomy procedures. Resin-
modified glass ionomer cements (RMGIC) and
some hydrophilic cements are excellent seal-
ants when combined with light-cured compos-
ite resins as permanent restorations, and
placed directly over DPC materials such as MTA
or other calcium silicate cements23-25.
Nowadays, instead of attending two
appoint- ments, patients can have their teeth
permanently restored during the pulpotomy
treatment (per- formed in one appointment).
The final restoration is placed with the aim of
sealing the pulpotomy material, and to further
defend the pulp from micro- leakage and
microbial challenges. Adhesive restora- tive
materials preserve the remaining tooth struc-
 materials. The most durable bond strengths are
achieved by using selective etching of enamel with 34% to
37% phosphoric acid, followed by two- step, two-bottle,
self-etching adhesive systems26,27. Following treatment, the
pulp health status must be assessed periodically to ensure
continued pulp vital- ity and the development of dentine
in the walls of the root canal and, in immature teeth,
apical closure. Even with modern techniques and
biocompat- ible materials, a statistically significant
difference in the successful outcome, might not be
achieved within a short follow-up period, compared
with the Cvek (1978) study. However, improved prog-
nosis is expected in the near future with today’s
techniques employing newer materials.

Case 1
A 24-year-old male patient presented with a com-
plicated enamel-dentine fracture in the maxillary left
central incisor due to a surfing accident that occured 4
hours previously; the coronal fragment was lost (Fig 1a).
The clinical examination revealed a tooth with a 0.5 mm
pulp exposure and enamel cracks cervically (Fig 1b);
there was no abnormal tooth mobility. A periapical
radiograph revealed a normal root maturation and no
obvious luxation injury (Fig 1c). The periodontal
probing depths were less than 3 mm. Teeth 11
(maxillary right central incisor), 21 (maxillary left central
incisor) and 22 (maxillary left lateral incisor) were vital
to cold testing with no injuries to tooth 11 or 22. The
treatment plan involved a Cvek pulpotomy and
restoration with composite resin.
After administration of a local anaesthetic and
dental dam application, a high-speed straight diamond
bur was used with water coolant to remove 2 mm of the
coronal pulp tissue. Bleed- ing was arrested within 20
seconds with a cotton pellet soaked in NaOCl. A hard-
setting calcium silicate cement was placed over the
wound fol- lowed by light-cured TheraCal LC resin (Fig
1d), and covered with a RMGIC (Vitrebond, 3M, St Paul,
MN, USA), which was also light-cured. The crown was
restored with G-aenial composite resin (GC Corporation,
Tokyo, Japan) (Fig 1e).
 Hamadani and Chandler A modern look at the Cvek pulpotomy

Fig 1a to f Case 1:
(a) Tooth 21 (maxil-
lary left central
incisor) coronal frac-
ture; the fractured
fragment was lost;
(b) The exposed
pulp of tooth
21 was evident;
(c) Radiograph of
tooth 21; (d)
Direct pulp
capping of the
wound; (e) The
crown of tooth 21
was restored with
composite resin;
a b c (f) Follow-up
radio- graph 1
week after the
accident.

d e

The management of traumatic dental (Fig 1f). A long-term follow-up is essential to detect any
injuries is challenging for a general dental
practitioner as it often presents as an
emergency. Emergency appointments may not
be available or a short time period
appointment might only be available and,
often the ideal materials are not at hand. In
particular, in relation to cases that present as
an emergency during the night or weekends,
with clinicians being forced to operate in
unfamiliar surroundings. In these cases, the
use of materials such as Biodentine or MTA
seem more appropri- ate since TheraCal only
releases a small amount of calcium28 and is
cytotoxic. The pH is also signifi- cantly lower
compared with other CH products29. A Cvek
pulpotomy was the appropriate treatment and
the composite resin restoration provided a
good immediate seal and aesthetic result. At
the 1 week follow-up, tooth 21 was symptom-
free and the pulp responded to cold testing
pathological changes. A
ceramic restoration (veneer
or crown) may be required
in the future to improve
aesthetics of the tooth21.

Case 2
An 11-year-old male patient
presented with a
complicated enamel-
dentine fracture to tooth 21
(maxillary left central incisor)
after suffering a fall 5 hours
previously; the fractured
coronal fragment was
retrieved. The clinical
examination revealed a 0.5
mm bleeding pulp exposure
(Fig 2a); there was no
abnormal tooth mobility. A
radiograph of the maxillary
left central incisor revealed
an imma- ture root with
open apex and no luxation
injury (Fig 2b). Periodontal
probing depths were less than
3 mm. Teeth 11 (maxillary
right central incisor), 21
(maxillary left central incisor)
and 22 (maxillary left lateral
incisor) were responsive to
cold, with no
Hamadani and Chandler A modern look at the Cvek pulpotomy

Fig 2a to e Case 2:
(a) Bleeding pulp
exposure of
tooth 21
(maxillary left
central incisor);
(b) A radiograph
of tooth 21
reveals
incomplete apical
develop- ment;
(c) Coronal
fractured frag-
ment of tooth 21
reattached; (d) A
6-month follow-up
radiograph of tooth
21; (e) A 1-year a b
follow-up radio-
graph of tooth 21;
the apical closure
was evident.

c d e

obvious injuries to tooth 11 or 22. The perfectly when replaced. As MTA has
treatment plan involved a direct pulp cap been shown to have
followed by bond- ing of the coronal fragment
with composite resin. After administration of a
local anaesthetic and dental dam application, a
high-speed diamond bur with water coolant was
used to remove 2 mm of the tooth structure
from the coronal fragment to provide space for
the DPC material. ProRoot MTA (Dentsply) was
placed over the exposure site and covered
with a RMGIC (Vitrebond, 3M) and was light-
cured. The coronal fragment was bonded to
the fractured site with RelyX Unicem (3M
ESPE, St Paul, MN, USA) and was also light-
cured (Fig 2c). After 6 months, when tested,
the tooth responded to cold and the periapical
tissues appeared normal on a check radiograph
(Fig 2d). The situation was
the same, at review, after 1 year (Fig 2e).
This case was also managed by a general
den- tal practitioner. Fortunately, the fractured
tooth fragment was retrieved and it fitted
very good outcomes, in this case, a DPC was per-
formed. However, it would have been more appro-
priate to carry out a Cvek pulpotomy, removing part of
the coronal pulp tissue. A small cavity was prepared
within the coronal fragment to provide space for the
DPC of MTA, which allowed good approximation. MTA
has been shown to discolour tooth structure; Biodentine
or an alternative, with similar properties, were not
available. However, the patient and his parents were
happy with the aesthetic result achieved.

References
1. Cvek M. A clinical report on partial pulpotomy and cap- ping with
calcium hydroxide in permanent incisors with complicated crown
fracture. J Endod 1978;4:232–237.
2. Haghgoo R, Abbasi F. A histopathological comparison of pulpotomy
with sodium hypochlorite and formocresol. Iran Endod J
2012;7:60–62.
3. Witherspoon DE. Vital pulp therapy with new materials: new
directions and treatment perspectives — permanent teeth.
Pediatr Dent 2008;30:220–224.
 Hamadani and Chandler A modern look at the Cvek pulpotomy

4. da Rosa WLO, Cocco AR, Silva TMD, et al. Current

17. Darvell BW, Wu RC. “MTA” – an hydraulic silicate
trends and future perspectives of dental pulp capping
cement: review update and setting reaction. Dent
materials: A systematic review. J Biomed Mater Res B,
Mater 2011;27:407–422.
App Biomater 2018;106:1358–1368.
18. Gandolfi MG, Van Landuyt K, Taddei P, Modena E, Van
5. Ricucci D, Loghin S, Lin LM, Spångberg LS, Tay FR. Is
Meerbeek B, Prati C. Environmental scanning electron
hard tissue formation in the dental pulp after the
microscopy connected with energy dispersive x-ray
death of the primary odontoblasts a regenerative or a
ana- lysis and Raman techniques to study ProRoot
reparative pro- cess? J Dent 2014;42:1156–1170.
mineral trioxide aggregate and calcium silicate
6. Mente J, Geletneky B, Ohle M, et al. Mineral trioxide
cements in wet conditions and in real time. J Endod
aggregate or calcium hydroxide direct pulp capping:
2010;36:851–857.
an analysis of the clinical treatment outcome. J Endod
19. Parirokh M, Torabinejad M, Dummer PMH. Mineral
2010; 36:806–813.
triox- ide aggregate and other bioactive endodontic
7. Kakehashi S, Stanley HR, Fitzgerald RJ. The effects of
cements: an updated overview – part I: vital pulp
surgical exposures of dental pulps in germ-free and
therapy. Int Endod J 2018;51:177–205.
con- ventional laboratory rats. Oral Surg Oral Med Oral
20. Zhang H, Pappen FG, Haapasalo M. Dentin enhances
Pathol 1965;20:340–349.
the antibacterial effect of mineral trioxide aggregate
8. Auschill TM, Arweiler NB, Hellwig E, Zamani-Alaei A,
and bioaggregate. J Endod 2009;35:221–224.
Scu- lean A. Success rate of direct pulp capping with
21. He LH, Purton DG, Swain MV. A suitable base material
calcium hydroxide [in German]. Schweiz Monatsschr
for composite resin restorations: zinc oxide eugenol. J
Zahnmed 2003;113:946–952.
Dent 2010;38:290–295.
9. Barthel CR, Rosenkranz B, Leuenberg A, Roulet JF.
22. Boeckh C, Schumacher E, Podbielski A, Haller B.
Pulp capping of carious exposures: treatment outcome
Antibac- terial activity of restorative dental
after 5 and 10 years: a retrospective study. J Endod
biomaterials in vitro. Caries Res 2002;36:101–107.
2000;26: 525–528.
23. Atabek D, Sillelioğlu H, Ölmez A. Bond strength of
10. Baume LJ, Holz J. Long term clinical assessment of
adhe- sive systems to mineral trioxide aggregate with
direct pulp capping. Int Dent J 1981;31:251–260.
different time intervals. J Endod 2012;38:1288–1292.
11. Hørsted P, Søndergaard B, Thylstrup A, El Attar K,
24. Eid AA, Komabayashi T, Watanabe E, Shiraishi T,
Fejer- skov O. A retrospective study of direct pulp
Watan- abe I. Characterization of the mineral trioxide
capping with calcium hydroxide compounds. Endod
aggregate- resin modified glass ionomer cement
Dent Traumatol 1985;1:29–34.
interface in different setting conditions. J Endod
12. Nair PN, Duncan HF, Pitt Ford TR, Luder HU. Histologi-
2012;38:1126–1129.
cal, ultrastructural and quantitative investigations on
25. Neelakantan P, Grotra D, Subbarao CV, Garcia-Godoy F.
the response of healthy human pulps to experimental
The shear bond strength of resin-based composite to
capping with Mineral Trioxide Aggregate: a randomized
white min- eral trioxide aggregate. J Am Dent Assoc
controlled trial. Int Endod J 2008;41:128–150.
2012;143:e40–e45.
13. Barnes IE, Kidd EA. Disappearing Dycal. Br Dent J 1979;
26. Cardoso MV, de Almeida Neves A, Mine A, et al.
147:111.
Current aspects on bonding effectiveness and stability
14. Poggio C, Arciola CR, Beltrami R, et al.
in adhesive dentistry. Aust Dental J 2011;56(suppl
Cytocompatibility and antibacterial properties of
1):31–44.
capping materials. Scienti- ficWorldJournal
27. Nikaido T, Weerasinghe DD, Waidyasekera K, Inoue G,
2014;2014:181945.
Foxton RM, Tagami J. Assessment of the nanostructure
15. Camilleri J, Montesin FE, Brady K, Sweeney R, Curtis
of acid-base resistant zone by the application of all-in-
RV, Ford TR. The constitution of mineral trioxide
one adhesive systems: Super dentin formation. Biomed
aggregate. Dent Mater 2005;21:297–303.
Mater Eng 2009;19:163–171.
16. Torabinejad M, Higa RK, McKendry DJ, Pitt Ford TR.
28. Camilleri J, Laurent P, About I. Hydration of
Dye leakage of four root end filling materials: effects
Biodentine, Theracal LC, and a prototype tricalcium
of blood contamination. J Endod 1994;20:159–163.
silicate-based dentin replacement material after pulp
capping in entire tooth cultures. J Endod
2014;40:1846–1854.
29. Subramaniam P, Konde S, Prashanth P. An in vitro
evalu- ation of pH variations in calcium hydroxide
liners. J Indian Soc Pedod Prev Dent 2006;24:144–145.
 Payman Hamadani, BDS Nicholas Chandler, BDS, MSc, PhD
Sir John Walsh Research Institute, Fac- Professor, Sir John Walsh Research Insti-
ulty of Dentistry, University of Otago, tute, Faculty of Dentistry, University of
Dunedin, New Zealand Otago, Dunedin, New Zealand

Payman Hamadani

Correspondence to:
Professor Nicholas Chandler, School of Dentistry, University of Otago, PO Box 56, Dunedin 9054, New Zealand. E-mail: