PEDIATRIC DENTISTRY V 39 I NO 1 JAN / FEB 17

Primary Tooth Vital Pulp Therapy: A Systematic Review and Meta-analysis

James A. Coll, DMD, MS' • N. Sue Seale, DDS, MSD3 • Kaaren Vargas, DDS, PhD3 • Abdullah A. Marghalani, BDS, MSD, DrPH4 • Shahad Al Shamali, BDM5
Laurel Graham, MLS6

Abstract: Purpose: This systematic review and meta-analysis assessed outcomes in prim ary teeth fo r the vital pulp therapy (VPT) options o f
indirect pulp therapy (IPT), direct pulp capping (DPC), and pulpotom y after a m inim um o f 12 m onths to determine whether one VPT was su­
perior. Methods: The following databases were searched from I960 to September 2016: MEDLINE, EMBASE, CENTRAL, EBSCO, ICTRP, Dissertation
abstracts, and grey literature fo r parallel and split-mouth randomized controlled trials o f at least 12 months durotion comparing the success o f
IPT, DPC, and pulpotom y in children with deep caries in prim ary teeth. Our prim ary outcome measure was overall success (combined clinical
and radiographic). Three authors determined the included RCTs, perform ed data extraction, and assessed the risk o f bias (ROB). Meta-analysis
and assignment o f quality o f evidence by Grading o f Recommendations Assessment, Development and Evaluation approach were done. Results:
Forty-one articles qualified fo r meta-analysis (six IPT, fo u r DPC, and 31 pulpotom y) from 322 screened articles. The 24-m onth success rates
were: IPT=94.4 percent, and the liner m aterial (calcium hydroxide [CH]/bonding agents) had no effect on success (P=0.88), based on a moderate
quality o f evidence; DP = 88.8 percent, and the capping agent (CH/alternate agent) did not affect success (P=0.56), based on a low quality o f
evidence. The combined success rate fo r all pulpotomies was 82.6 percent based on 1,022 teeth. Mineral trioxide aggregate (MTA) (89.6 percent)
and formocresol (EC) (85.0 percent) success rates were the highest o f all pulpotom y types and were n o t significantly different (P=0.15), with
a high quality o f evidence. MTA’s success rate (92.2 percent) was higher than ferric sulfate (FS) (793 percent) and approached significance
(P=0.06), while FS’s success rate (84.8 percent) was not significantly different from FC (87.1 percent), both with a moderate quality o f evidence.
MTA and FC success rates were significantly better than CH (P=0.0001), with a moderate quality o f evidence. A t 18 months, sodium hypochlorite
(NaOCI) success rate was significantly less than FC (P=0.0l) with a low quality o f evidence. Conclusions: The highest level o f success and quality
o f evidence supported IPT and the pulpotom y techniques o f MTA and FC fo r the treatm ent o f deep caries in prim ary teeth after 24-months.
DPC showed similar success rates to IPT and MTA o r FC pulpotomy, b u t the quality o f the evidence was lower. Systematic Review Registration
Number: PROSPERO 2015: CRD42015006942. (Pediatr Dent 2017;39(1):16-27.E15-E95) Received September 28, 2016 I Accepted December 15, 2016

KEYWORDS: PRIMARY TEETH, PULP THERAPY, SYSTEMATIC REVIEW, META-ANALYSIS

Dental caries is the most common chronic disease of child­ ture since the mid 1800’s.4,5 However, a number of studies of
hood.1When caries is untreated it can endanger the vitality of IPT in primary teeth have been published with encouraging
the tooth, causing infection, pain, abscess, and premature tooth results.^10
loss; thus, early intervention is very important for the child’s DPC is a technique in which the pulp is covered with a
overall health and development.2 biocompatible material when caries excavation causes a pin­
Currently, there are three vital pulp therapy (VPT) options point pulp exposure.11 Past reports of DPC in primary teeth
for treatment of deep dental caries approximating the pulp in have shown limited success;12,13therefore, DPC has had lim­
primary teeth (1) indirect pulp treatment (IPT), also known ited acceptance as a technique for management of carious pulp
as indirect pulp cap; (2) direct pulp cap (DPC); and (3) pulpo­ exposures in the primary dentition.
tomy. 2,3 Pulpotomy is a procedure used when the excavation of caries
IPT is a procedure that leaves the deepest caries adjacent in primary teeth produces a carious pulp exposure. In this tech­
to the pulp undisturbed in an effort to avoid a pulp exposure. nique, the entire coronal pulp is removed, hemostasis of the
This caries-affected dentin is covered with a biocompatible radicular pulp is achieved, and the remaining radicular pulp is
material to produce a biological seal.2-’The use of IPT remains treated with one of several different medicaments.2,3 Published
controversial even though it has been described in the litera- studies of this procedure have been reported since the early
1900’s,14 and currently pulpotomy is the most frequently used
VPT technique for deep dental caries in primary teeth.15
‘Dr. Coll is a clinical professor and *Dr. Marghalani is a pediatric dental fellow, Divi­
sion of Pediatric Dentistry, both at the University of Maryland Dental School, Balti­
more, Md., USA. -Or. Seale is a Regents Professor, Department of Pediatric Dentistry, ABBREVIATION KEY.
Texas A&M College of Dentistry, Dallas, Texas, USA. 3Dr. Vargas is in private practice,
Corridor Kids Pediatric Dentistry, North Liberty, Iowa, USA. sDr. Al Shamali is a VPT: Vital pulp therapy. IPT: Indirect pulp treatment. DPC: Direct
Resident in Pediatric Dentistry. Department of Pediatric Dentistry, University of Illinois pulp cap. RCTs: Randomized controlled trials. RR: Relative risks.
at Chicago College of Dentistry; and 6Ms. Graham is Evidence-Based Dentistry Manager, Cl: Confidence intervals. NNT: Number need to treat. ROB: Risk-
American Academy of Pediatric Dentistry, both in Chicago, III., USA. of-bias. GRADE: Grading of Recommendations Assessment, De­
Correspond with Dr. Seale at velopment, and Evaluation. FC: Formocresol. MTA: Mineral trioxide
Supplemental material available aggregate. FS: Ferric sulfate. NaOCL: Sodium hypochlorite. CH:
in the online version. Calcium hydroxide.

16 VITAL PULP THERAPY


Many studies have been conducted to determine the effi­
cacy of IPT, DPC, and pulpotomy. Investigations vary in design,
techniques, materials used, and diagnostics for determining
and reporting outcomes, all of which affect the quality of the
evidence produced.6' 10,12'13-16"18 Therefore, standardization of
methodology has been critical to analyzing VPT studies; in
recent years, as the number of randomized controlled clinical
trials (RCTs) conducted and published has increased, reporting
quality has improved.
VPT is so frequently used in the management and mainte­
nance of the primary dentition that an analysis of the existing
literature is imperative. Previous systematic reviews and meta­
analyses have compared one or more types of vital pulp ther­
apy;19-21 however, none have included all the three vital pulp
therapy options (IPT, DPC, and pulpotomy) or compared
medicaments for these treatment options.19-21
The purpose of this investigation was to perform a systema­
tic review and meta-analysis of VPT in cariously involved vital
primary teeth after a minimum of 12 months to determine the
overall (combined clinical and radiographic) success rates and
whether one VPT was superior. In addition, we evaluated VPT
moderators/factors that may have had effects on outcomes such
as method of isolation, type of final restoration, and number
of appointments to treat.
Methods
In order to enhance the transparency, a protocol was published22
and is available on open access at: “http://www.ingentaconnect.
com/content/aapd/pd/2015/00000037/00000005/art00002”.
Search strategy. One of the authors (LG) searched the
following databases from 1960 to September, 2016: MEDLINE
(PubM ed) 1960; EMBASE; Cochrane C entral Register of
Controlled Trials (CENTRAL); EBSCO— Dentistry and Oral
Sciences Source; IC TR P (trials database); Dissertation ab­
stracts; and the grey literature. The search used MeSH terms
and keywords to find both published and unpublished studies.
Note that there is not a MeSH term for indirect pulp therapy;
the phrase pulp therapy was used to retrieve IPT studies. We
searched relevant journals from 1990 onward, since that date
was the earliest we expected to find RCTs. See Electronic Ap­
pendix: Section 1 for search strategy and a list of journals in­
dividually searched. We also used an objective search strategy,
which is a variation of the method listed in Routine Develop­
ment of Objectively Derived Search Strategies.23 The resulting
search strategy retrieved all vital pulp therapy included studies
cited by the recent Cochrane pulp therapy review.19
Titles and abstracts were screened by three reviewers (JAC,
NSS, KV) to identify studies for inclusion in the systematic
review. A member of the work group provided translations for
Spanish and Portuguese studies. W hen a paper was in a non-
English language other than Spanish, or Portuguese and needed
further review, a translation was requested.
If a study’s abstract was unclear, the full report was accessed
to determine eligibility for inclusion. If a study published re­
sults for different time frames, data from all of the publications
were considered for inclusion. All papers were reviewed in du­
plicate by the authors (JAC, NSS, KV) to determine inclusion
as described in the Selection Criteria.
Selection criteria. Inclusion and exclusion criteria were
based on the population, intervention, comparison, outcomes,
and study design (PICOS) method.
P EDIATRIC D E N TIS T R Y V 3 9 I NO 1 JAN I FEB 17
The population was defined as healthy pediatric patients
who required vital pulp therapy for deep caries in primary teeth
including molars, incisors, and canines, and the tooth was our
unit of analysis. The intervention was any of the three types of
VPT (IPT, DPC, and pulpotomy of any type), and the compar­
ison was to any other VPT. Pulpal treatments as a result of
non-carious pulp exposures were excluded. Outcomes were
reported as overall success, determined as simultaneous clinical
and radiographic success, after a minimum of 12 months. Study
design was limited to RCTs.
Data extraction. Two of the three reviewers (JC, SS, KV)
independently performed the data extraction and risk-of-bias
assessment for the included articles using a standardized tem­
plate. Forty-nine fields were extracted from each study when
possible (Electronic Appendix: Section la). If there were ques­
tions regarding data, attempts were made to contact the study’s
authors. Disagreements were resolved through discussion by the
reviewers. For RCTs with more than two arms, only relevant
arms were considered. RCTs with split-m outh designs were
combined with parallel designs for possible subgroup analysis
and meta-regression analysis.
Data synthesis. Meta-analysis were done by one author
(AAM) using systematic review software programs Review
Manager (RevMan) Version 5.2.1 and Comprehensive Meta-
Analysis (CMA) 3.0 software (Biostat, Englewood, N.J., USA).
We used the random effects models, inverse variance method,
to obtain pooled relative risks (RR) and 95 percent confidence
intervals (Cl) for overall success. The number needed to treat
(NNT) for each outcome was determined by reciprocating the
pooled risk difference. We reported the overall success as a per­
centage with 95% Cl by combining success proportions in
each and both comparison groups, provided there was no signi­
ficant difference between the two groups, using random-effect
models considering within and between-trials variance. The
heterogeneity was determined by using the I2 statistic and the
Cochrane test for heterogeneity, with P less than 0.1 considered
to be statistically significant. We assessed publication bias if
the number of included RCTs exceeded 10. We conducted
sensitivity analysis by including high risk of bias or industry
funded trials in the meta-analyses.
Outcomes. Primary Outcomes. We defined overall success
as only those teeth that showed both clinical (absence of spon­
taneous pain, soft tissue pathology, and pathologic mobility)
and radiographic success (absence of internal/external, furca­
tion, or periapical radiolucency) simultaneously in studies with
a follow-up of at least 12 months.
Variation In Primary Outcome Measures. Rules were devel­
oped to recalculate and standardize outcome data for some
studies, because determ ination of success and failure varied
according to outcome measurement. Data collection for exfoli­
ated teeth varied among studies. Some studies considered nor­
mal, uneventfully exfoliated teeth to be dropouts and removed
them from the calculations for success.17-24'26 Some studies
reported a failure in one time frame but did not carry that
failure forward to the next time fram e.17-24-25 Some studies
reported an overall success separately as various combinations
of clinical and radiographic success.6,10,11'13-18-27'31 The remaining
studies rated radiographic success as the overall success. In a
few studies internal root resorption was rated as a success,32'34
and one study did not consider excess mobility of a treated
tooth as a failure.25 Studies included in the meta-analysis had
data recalculated when necessary by two authors using these
VPT Standardization Rules (JAC, SAS).
V ITA L PULP TH E R A P Y 17
P E D IA TR IC D E N TIS T R Y V 39 I NO 1 JAN / FEB 17

VPT Standardization Rules. Overall VPT success was stand­ assigned a score of four or six, with six=18 months and four
ardized by applying the following rules: = 12 months. Magnitude o f the effect was determined by the
1. Following a VPT, if a tooth exfoliated in less than six overall success of the medicaments in each arm of the meta­
months, it was counted as a failure. analysis. The overall quality of evidence was assessed for each
2. If a tooth exfoliated greater than six months after VPT, important outcome as high, moderate, low, or very low.
it was always counted as a success in all future time
frames. Results
3. Once a tooth’s VPT failed in a time frame, it was cal­ Description o f studies
culated as a failure in all future time frames. The table describing the characteristics of the 41 studies in­
4. Contained internal resorption and excess mobility were cluded in the meta-analysis is in the Electronic Appendix:
counted as failures in any time frame. Section 3. Initial searches from all sources identified 2,204
5. A drop out in any time frame was removed from that references on pulp therapy in primary teeth, which yielded 926
time frame’s denominator and all future time frames in non-duplicate titles. Abstracts from 322 were reviewed and
calculating success. 148 excluded, because they were non-RCTs or studies of non-
6. Overall success was defined as only those teeth that vital treatment, leaving 174 for full text assessment. Following
showed both clinical and radiographic success simul­ full-text assessment for eligibility, 101 articles were excluded
taneously. with reasons (wrong study design, wrong population, and/
or wrong follow-up) leaving 71 articles for data extraction.
The definition of overall VPT success was revised since Following review, 41 articles were determined to be low or
the protocol was published, and this revised definition was unclear ROB and qualified for inclusion in one or more
used in reporting data. meta-analyses (Figure 1). The 30 not eligible for meta-analysis
Secondary Outcomes. The secondary outcomes of pain, soft included 21 that were determ ined to be unique vital pulp
tissue pathology, pathologic mobility, furcation and periapi­ treatm ent comparisons that could not be compared to any
cal radioluocency, and external/internal root resorption were other study12,18-27,3234'38'51,52-53 (Electronic Appendix: Section 4).
not evaluated, because studies did not describe these o u t­ The 9 that were rated as high bias54'62 (Electronic Appendix:
comes in sufficient detail to allow consistent data extraction Section 3) were used in meta-analysis as part of the sensitivity
among studies. analyses.
Outcome moderators/factors. T he hypothetical outcome The 41 articles in the meta-analysis actually represented
moderators/factors evaluated for their potential significant effect 39 separate studies, because four were two different studies
on success using subgroup and meta-regression analyses in­ reported at two different time intervals.7'5,28 These 39 studies
cluded method of tooth isolation, type of final restoration, ti­ included nine split-mouth designs and 30 parallel-arm designs
ming of final restoration placement, hemostasis method, type with 3,709 randomized primary teeth in 2,078 children ranging
of IPT, DPC, and pulpotomy base material, and calcific meta­
morphosis.
Risk o f bias assessment. The Cochrane Collaboration’s
risk-of-bias assessment tool was used to create an overall assess­
ment of the risk-of-bias (ROB) for each study based on key
domains (Electronic Appendix: Section 2).35 A low ROB was
when all of the key domains of bias were judged to have low
risk. An unclear ROB had at least one key domain judged as
unclear. The high ROB had at least one key domain judged
to have a high risk of bias. Only studies with low and unclear
ROB were used in the meta-analysis presented in this paper.
Grading. The Grading of Recommendations Assessment,
Development, and Evaluation (GRADE) system was used to
judge the quality of evidence of each outcome included in the
meta-analysis.36 The criteria used included ROB/study limit­
ations, consistency of results, precision, importance, and mag­
nitude of the effect. ROB was determined using the Cochrane
Risk of Bias Tool and GRADE guidance on overall ROB. In
the meta-analyses Consistency was judged based on the hetero­
geneity (I2) of the studies in the meta-analysis and assigned
as Not Serious (Low): I2=0-30 percent, Serious (Moderate):
I2=35-65 percent, and Very Serious (High): I2=>75 percent.37
Precision was based on the total sample size in each arm of
the meta-analysis. Sample size imprecision was assigned as
Not Serious (greater than 125 total teeth in each arm), Serious
due to small sample size (65-110 total teeth in each arm),
and Very Serious due to very small sample size (less than 50
total teeth in each arm). Importance was judged as Critical or
Im portant based on length o f follow-up in m onths in the
meta-analysis. Critical was assigned a score of eight or nine
with nine=36 months and eight=24 months; Im portant was Figure 1. Flow chart.

18 V ITA L PULP TH ER APY

 PEDIATRIC DENTISTRY V 39 l NO 1 JAN I FEB 17

in age from 2.3 to 12.5 years. Studies included in meta-analysis O f the 41 articles used in the meta-analysis, six were 1PT
by country of origin are reported in the Electronic Appendix: articles comparing different IPT liners’ success rates (Electronic
Section 5. Appendix: Section 6, Table le), four were DPC articles com­
The 71 trials that underwent data extraction were pub­ paring different pulp capping materials’ success rates, (Electro­
lished between 1991 and 2016 with the majority (72 percent) nic Appendix: Section 6, Table 2e) and 31 were pulpotomy
published between 2005 and 2012. They were all single-center articles comparing formocresol (FC), mineral trioxide aggregate
trials except one, which was a multicenter trial.16 These trials (MTA), ferric sulfate (FS), sodium hypochlorite (NaOCL), laser,
were conducted in pediatric dentistry departments of a uni­ calcium hydroxide (CH), and tricalcium silicate (Biodentine)
versity or hospital. All trials were conducted by dentists, success rates. (Electronic Appendix: Section 6, Table 3e).
residents supervised by pediatric dental faculty, or fifth year The ROB assessment table for the 71 included studies
dental students. from which data were extracted is in the Electronic Appen­
dix: Section 7, and only four studies met the criteria to be
classified at low ROB.24,34,63,64

Overall success, low and unclear risk of bias trials, 24 months

Mo CH CH R isk R atio R is k R atio
Study o r S ub g rou p E ven ts T o ta l E ven ts T o ta l W e ig h t IV. R andom , 95% Cl IV, R andom . 95% Cl

Buyukgural et al., 2008 180 180 60 60 97.4% 1.00 [0.98,1.02)

C asag ra n de et al., 2008 14 16 13 15 0.8% 1.01 [ 0 . 7 7 , 1 . 3 2 ) -------------------------------
F a ls te re ta l., 2002 24 25 20 23 1.8% 1.10 [0.9 2,1 .32 ) ---------------------

T o ta l (95% Cl) 221 98 100.0% 1 .0 0 1 0 .9 8 ,1 .0 3 ] i ■

Total events 218 93
H eterogeneity: Tau3 = 0.00; C h P = 1 18. d f = 2 (P = 0 56), Is = 0% I------------------ 1------------------- -----------------------h - H
T e s t for overall e ffe c t Z = 0.15 (P = 0.88)
0.5 0 7 1.5 2
Favors CH Favors N o CH

R D = 0 .0 0 (-0 .0 2 , 0 .0 3 ), N N T = c a n n o t b e c a lc u la te d , n o t s ig n ific a n t.

Figure 2. Forest plot o f IPC success at 24 months.

Overall success, low and unclear risk of bias trials, 24 months *

A lte rn a te CH R isk Ratio R is k R atio
Study o r Subgroup Events T otal Events T otal W e ig h t fV, R andom , 95% a IV, R andom , 95% Cl

A m in ab a di e t al., 2010 51 60 36 60 24 1% 1.42 [1.12,1.79)

D e m ira n d C ehreli, 2007 72 80 20 20 37 5% 0 92 [0 8 3 .1 01)
T u na and O lmez, 2008 22 22 20 20 384% 1 0 0 (0 .9 1 ,1 091

T o ta l (95% Cl) 162 100 100.0% 1.05 [0 .8 9 ,1 .2 5 ]

T o ta l events 145 76
Heterogeneity: Taua = 0.02; C h P r 11 44, d f= 2 (P = 0 003); P = 83% j— ■ H -------------------------------------1—
0.7 1.5 2
T e st fo r overall effect: Z = 0.59 (P = 0.56) 05
Favors C H Favors Alternate

R D = 0 .0 4 (-0 .1 3 , 0 .2 1 ). N N T = 2 5 , n o t s ig n ific a n t

Figure 3. Forest plot o f D PC success at 24 months.

Overall success, low and unclear risk of bias trials, 24 months *

MTA FC R isk Ratio R isk Ratio
Study o r Subgroup Events T otal Events Total W e ig ht IV, Random . 95% a IV. Random . 95% Cl

E rd e m e ta l., 2011 24 25 23 25 163% 1.04 10 9 1.1 .2 0 )

Fernandez et al., 2013 20 20 24 25 23.3% 1.04 [0.92,1.16)
Moretii et al., 2008 14 14 15 15 19 2% 1.00 [0.88.1.14)
N o orollah ia n 2008 17 19 18 18 10.0% 0.90 |0 75.1.08)
S o n m e z e ta l, 2008 10 15 10 13 1.6% 0 87 )0 54, 1.38)
S ub ra m a nlam et al., 2009 19 20 17 20 7.5% 1.1 2 (0 9 1,1 .3 8 )
S ushynskt et al., 2012 58 65 49 66 119% 1 20 11.02,1 42]
Y ild irim et al., 2016 57 63 27 33 10.2% 1.11 (0 92,1 32]

T o ta l (95% Cl) 241 215 100.0% 1 .0 4 (0 .9 8 ,1 .1 1 ]

T otal events 219 183
Heterogeneity: T a u *= 0 00. C h i*= 7 27, d f= 7 (P = 0.40), P = 4% -t- I
T est for overall effect: Z = 1.44 (P = 0.15) 1.5 2
Favors FC Favors MTA

R D = 0 .0 4 (-0 .0 1 , 0 .1 0 ) N N T = 2 5 , n o t s ig n ific a n t

Figure 4. Forest Plot for Pulpotomy Success for Formocresol (FC) versus M ineral Trioxide Aggregate (MTA)
at 24 months.

* CH=Calcium Hydroxide. RD=Risk Difference. N N T=N um ber Needed to Treat. CI=Confidence Interval.
IV, Random=inverse variance random-effect model. df=degrees of freedom.

VITAL PULP THERAPY 19

P E D IA TR IC D EN TIS T R Y V 3 9 / NO 1 JAN I FEB 17
Meta-Analysis Results Comparing Different Pulp Treat­
ment Interventions On Success Outcomes
IPT. Six articles6’10,28 compared IPT success using CH liners
versus bonding agent liners. The 24-month overall success rate
of IPT irrespective of the liner was 94.4 percent (95% Cl:
84.9, 98.0 percent). The meta-analysis showed the liner had
no effect on IPT success (P=0.88) (RR 1.00 95% Cl: 0.98,
1.03) (Figure 2). At 48 months, the overall success rate of IPT
decreased to 83.4 percent (95% Cl: 72.9 percent, 90.4 per­
cent), with the meta-analysis showing no significant difference
between CH or bonding agent/liners (RR 1.10 95% Cl: 0.92,
1.32) (/J=0.31). The quality of the evidence for liners was best
at 24 months, and was assessed as moderate because of small
sample size issues. At 48-months the quality of evidence was
assessed as low due to the very small sample size issues. Forest
plots for IPT for all follow-up times (12, 18, 24, and 48 months)
are included in the Electronic Appendix: Section 8, and the
Summary of Findings for IPT are included in the Electronic
Appendix: Section 8, Table le.
Direct Pulp Cap. Three o f the four DPC articles com­
pared CH versus alternative direct capping agents after 24-
months (dentin bonding agents11, MTA29, and FC31). The
24-m onth overall success rate of D PC irrespective of the
capping agent was 88.8 percent (95% Cl: 73.3, 95.8), and
the meta-analysis showed the capping agent had no effect on
success (RR 1.05 95% Cl: 0.89, 1.25) CP=0.56). However, the
quality of the evidence for whether DPC capping agent affected
success at 24 months was assessed as very low because of the
high degree of heterogeneity in the studies (I2= 83%) and small
sample size (Figure 3). Forest plots for DPC for all follow-up
times (12, 18, and 24, months) are included in the Electronic
Appendix: Section 9, and the Summary of Findings for DPC
are included in the Electronic Appendix: Section 9, Table le.
Pulpotomy. When the results were pooled for all 31 pulpo-
tomy articles used for the meta-analysis, the overall success
rate from the MTA, FC, FS, NaOCL, CH, and laser 24-month
studies was 82.6 percent (95% Cl: 75.8, 87.8 percent ) based
on 1,022 teeth. MTA and FC 24-m onth success rates were
the highest of all pulpotom y types in this time frame and
were not significantly different (/>=0.15). MTA’s success rate
was 89.6 percent (95% Cl: 82.5, 94.0), and FC’s was 85.0
percent (95% Cl: 76.3, 91.0). At 24 months, MTA was sig­
nificantly better than FS based on the N N T o f nine, indi­
catin g th a t after d o in g n ine M TA p u lp o to m ie s, one
failure would be prevented than if FS was used. MTA, FC,
and FS success rates were all significantly better than CH
at 24 m onths (7>=<0.001). At 18 m onths, N aO C l’s success
rate was significantly less than FC (P=0.01), but other pulp­
otom y agents’ success rates did not differ statistically (FS
vs. laser; FS vs. NaOCL; and CH vs. laser). At 12 months,
p u lpoto m y success rates for FS vs. laser and MTA vs.
Biodentine were not significantly different. The Summary
of Findings for all the pulpotomy studies at their maximum
length of follow-up are included in the Electronic Appen­
dix: Section 10.
FC vs. M TA p u lp o to m y (24-m onths). E ight articles
comparing FC to MTA had a follow-up of 24 months.16,30,65'70
The FC overall success rate in these articles was 85.0 percent
(95% Cl: 76.3, 91.0), while MTA’s was 89.6 percent (95%
Cl: 82.5, 94.0), with the meta-analysis favoring neither agent’s
success (RR 1.04 95% Cl: 0.98, 1.11) (/>=0.15). The quality of the
evidence for this outcome at 24-months was high due to no
GRADE issues (Figure 4). Forest plots for FC vs. MTA pulp­
20 V IT A L PULP TH E R A P Y
otomies for all follow-up times (12, 18, and 24, months) are
included in Electronic Appendix: Section 10a.
FC vs. FS pulpotom y (24-months). Four articles com ­
paring FC to FS had a follow-up of 24 m onths.17,65'66,68 The
FC overall success rate was 87.1 percent (95% Cl: 78.2, 92.7
percent), and FS’s was 84.8 percent (95% Cl: 76.2, 90.6 per­
cent), with the meta-analysis favoring neither agent’s success
(RR 1.02 95% Cl: 0.93, 1.13) (/>=0.65). The quality of the
evidence for this outcome at 24-months was moderate due to
small sample sizes. Forest plots for FC vs. FS pulpotomies for
all follow-up times (12, 18, and 24, months) are included in
Electronic Appendix: Section 10b.
FC vs. CH pulpotom y (24-months). Four articles com ­
paring FC to C H had a follow-up o f 24 m o n th s.17,33,67,68
The FC overall success rate was 79.0 percent (95% Cl: 57.7,
91.2 percent) and C H ’s was 41.4 percent (95% Cl: 26.5, 58.1
percent), with the meta-analysis indicating FC was signifi­
cantly b etter than C H (RR 1.76 95% C l: 1.40, 2.23)
(A><0.001). The N N T of three was significant. The quality of
the evidence for this outcome at 24-months was moderate due
to small sample sizes. Forest plots for FC vs. CH pulpoto­
mies for all follow-up times (12, 18, and 24 months) are
included in Electronic Appendix: Section 10c.
M TA vs. CH pulpotomy (24-months). Three articles com­
paring MTA to CH had a follow-up of 24 months.67,68,74 The
MTA overall success rate was 89.0 percent (95% Cl: 59.6,
97.8 percent), and CH was 46.0 percent (95% Cl: 35.0, 57-3
percent), with the meta-analysis indicating MTA was signifi­
cantly b etter than C H (RR 1.96 95% C l: 1.52, 2.53)
(TcO.OOOl). The N N T of three was significant. The quality of
the evidence for this outcome at 24-months was moderate due
to small sample sizes (Figure 5). Forest plots for MTA vs. CH
pulpotomies for all follow-up times (12, 18, and 24 months) are
included in Electronic Appendix: Section lOd.
FS vs. CH pulpotomy (24-months). Two articles comparing
FS to CH had a follow-up of 24 m onths.17,68 The FS overall
success rate was 82.1 percent (95% Cl: 68.2, 90.7 percent), and
CH was 52.8 percent (95% Cl: 39.5, 65.8 percent), with the
meta-analysis indicating FS was significantly better than CH.
(RR 1.57 95% Cl: 1.19, 2.06) (/Ml.OOl). The N N T of four was
significant. The quality of the evidence for this outcome at 24-
months was low due to very small sample sizes. Forest plots
for FS vs. CH pulpotomies for all follow-up times (12, 18, and
24 months) are included in Electronic Appendix: Section lOe.
MTA vs. FS pulpotomy (24-months). Four articles compa­
ring MTA to FS had a follow-up of 24 m onths.65,66,68,7'’ The
MTA overall success rate was 92.2 percent (95% Cl: 70.7,
98.3 percent), and FS’s was 79.3 percent (95% Cl: 68.0, 87.4
percent), with the meta-analysis nearing significance (/->=0.06)
favoring MTA (RR 1.13 95% Cl: 1.00, 1.29). The N N T of
nine was significant (Figure 6). The quality of the evidence for
this outcome at 24-months was moderate due to small sample
sizes. Forest plots for MTA vs. FS pulpotomies for all follow­
up times (12 and 24 months) are included in Electronic Ap­
pendix: Section lOf.
FC vs. NaOClpulpotomy (18-months). Two articles com­
paring FC to N aO C l had a m axim um follow -up o f 18
months.63,65 The FC overall success rate was 98.1 percent (95%
Cl: 97.6, 99.7 percent) and NaOCl’s was 82.9 percent (95%
Cl: 68.3, 91.6%), with the meta-analysis indicating FC was
significantly better than N aO C l (RR 1.20 95% Cl: 1.04,
1.40) (7A0.01). The N N T of six was significant (Figure 7).
The quality of the evidence for this outcome at 18-months was
 P E D IA TR IC D E N TIS T R Y V 39 / NO 1 JAN I FEB 17

Overall success, low and unclear risk of bias trials, 24 months *

MTA CH Risk Ratio Risk Ratio

Study or Subgroup Events Total Events Total Weight IV. Random, 95% Cl IV. Random, 95% Cl

Celik et al , 2012 82 87 22 47 67 6% 2 01 11.48.2.74] ........■

Moretti et al.,2008 14 14 6 14 18 8% 2 23 |1 24. 4.01]
Sonmez et al., 2008 10 15 6 13 13 7% 1 44 [0.73, 2 87]

Total (95% Cl) 116 74 100.0% 1.96 [1-52.2.53]

Total events 106 34
—
Heterogeneity Tau* = 0 00; Chi* = 0 97, df = 2 (P = 0 61), P=Q%
Test for overall effect Z = 5 20 (P < 0.00001)
oh oh t!s 1
Favors CH Favors MTA

RD = 0 .4 6 (0.30, 0.61), N N T = 3 (2, 4 ) sig nifican t

Figure 5. Forest Plot for Pulpotomy Success for FC vs. MTA at 24 months.

Overall success, low and unclear risk of bias trials, 24 months *

MTA FS Risk Ratio Risk Ratio
Study or Subgroup Events Total Events Total Werght IV. Random. 95% Cl IV. Random, 95% Cl

Doyle et al . 2010 45 47 34 46 30 7% 1 30(1.08,1 55]

Erdem et a l. 2011 24 25 23 25 41.1%
Fernandez et a l, 2013 20 20 12 14 21 5%
Sonmez et a l. 2008 10 15 11 15 67%

Total (95% Cl) 107 100 100.0% 1.13 [1.00,1.29J

Total events 99 80
Heterogeneity Tau* = 0 01. Chi* = 4 31, df= 3 (P = 0 23). I1•= 30%
05 07 15
Test for overall effect Z = 1 9 2 (P =; 0 06)
Favors FS Favors MTA

RD = 0.11 (0.00, 0.22), N N T = 9 sig nifican t

Figure 6. Forest Plot for Pulpotomy Success for MTA vs. FS at 24 months.

Overall success, low and unclear risk of bias trials, 18 months *

FC NaOCL Risk Ratio Risk Ratio

Study or Subgroup Events Total Events Total Weight IV, Random, 95% Cl IV. Random, 95% Cl

Farsi et a l, 2015 25 25 20 24 59 8% 1 20(0 99. 1 45]

Fernandez el a l, 2013 25 25 14 17 40 2% 1 22 [0 96. 1 54]

Total (95% Cl) 50 41 100.0% 1.20 [1.04,1.40)

Total events 50 34
Heterogeneity Tau*=0 00. Chl*= 001. df = 1 (p = 0 91);P=0%
Test for overall effect Z = 2 47 (P = 0 01)
Favors NaOCL Favors FC

RD = 0.17 (0.05, 0.29). NN T = 6 (4, 2 0 ) sig nifican t

Figure 7. Forest Plot for Pulpotomy Success for FC vs. N aO C l at 18 months.

* CH=Calcium Hydroxide. RD=Risk Difference. N N T=N um ber Needed to Treat.

CI=Confidence Interval. IV, Random=inverse variance random-effect model. df=degrees o f freedom.

moderate due to small sample sizes. Forest plots for FC vs.
N aO C L pulpotom ies for all follow-up times (12 and 18
months) are included in Electronic Appendix: Section 1Og.
FC vs. Laser; FS vs. NaOCl; CH vs. Laser pulpotomy (18-
months). The quality of the evidence for the outcomes of these
agent comparisons at 18-months was low due to small sample
sizes. Two articles compared FC to laser, 17,24and the m eta­
analysis favored neither type of pulpotomy (RR 1.14 95% Cl:
0.91, 1.43) {P=0.27). Forest plots for FC vs. Laser pulpotomies
for all follow-up times (12 and 18 months) are included in
Electronic Appendix: Section lOh. Two articles compared FS
to NaOCL,63,65 and the meta-analysis favored neither type of
pulpotom y (RR 0.99 95% Cl: 0.85, 1.16) (T^O.88) Forest
plots for FS vs. NaOCl pulpotomies for all follow-up times (12
and 18 months) are included in Electronic Appendix: Section
lOi. Two articles compared C H to laser,17,24 and the meta­
analysis favored neither type of pulpotomy (RR=1.07 95% Cl:
0.91, 1.25) (T^O.41). Forest plots for CH vs. Laser pulpotomies
for 12 and 18 months follow-up are included in Electronic
Appendix: Section lOj.
FS vs. Laser; MTA vs. Biodentine Pidpotomy (12-months).
The quality of the evidence for these outcomes at 12-months
was very low due to small sample sizes and short follow-up
duration. Two articles compared FS to laser,17,76 and the meta­
analysis favored neither type of pulpotomy (RR 1.06 95% Cl:
0.94, 1.19) (/L0.34). Forest plots for FS vs. Laser pulpotomies
for 12 months follow-up are included in Electronic Appendix:
Section 10k. Two articles compared MTA to Biodentine,64,77
and the meta-analysis favored neither pulpotomy (RR=1.01
95% Cl: 0.94, 1.09) (^ G .8 3 ). Forest plots for Biodentione
vs. MTA pulpotomies for 12 months follow-up are included
in Electronic Appendix: Section 10 L.

V ITA L PULP T H E R A P Y 21
PE D IA TR IC D E N TIS T R Y V 3 9 / NO 1 JAN I FEB 17
Outcome moderators/factors results
We had 21 trials with 51 arms with unclear/low risk of bias
that lasted for 24 months, and all trials involved molars. Mean­
ingful testing was not possible for the majority of the factors/
modifiers due to a low number of trials in each comparison,
incomplete data reporting, missing data, or unbalanced event
distribution. The three exceptions were for type of final restora­
tion (stainless steel crowns (SSC) vs. fillings), study design
(split-mouth design vs. parallel design) and rubber dam use (yes
vs. no).
For the type of final restoration, we compared the SSC
success rate of 82.4 percent with the intracoronal restoration
success rate of 84.2 percent, and there was no significant differ­
ence at 24 m onths using the sub-group analysis (/VO.697)
and meta-regression analysis (/VO.6939). In addition, we com­
pared success rates in split-m outh and parallel arm studies,
and there were no significant differences between them using
the sub-group analysis (/VO.746) and m eta-regression
analysis (/>=0.7467). For use or non-use of rubber dam, the
arms from 41 studies using rubber dam were compared to the
arms of six that did not; pulp therapy was more successful
when rubber dam was used, but not significantly more, so at
24 months using the sub-group analysis (/VO.334) and meta­
regression analysis (/VO.3299).
Sensitivity analysis
We conducted sensitivity analysis by including high ROB trials
in the meta-analysis, and the forest plots are included in the
Electronic Appendix: Section 10a-l. The inclusion of high
risk of bias trials did not change the RR more than five percent.
In one analysis, MTA vs. FC at 24 months, the RR became
statistically significant (/VO.01) when high ROB trials were
included (Electronic Appendix: Section 10a). We also intended
to include industry-funded trials in the sensitivity analysis, but
there were none.
Discussion
The results of this systematic review and meta-analysis showed
variable success rates and levels of evidence for the treatment
of deep dental caries in primary teeth. The 24-month overall
success rates for IPT, DPC, and pulpotomy were 94.4 percent,
88.8 percent, and 82.6 percent respectively. They were all viable
options for treatment of deep caries in primary teeth, but IPT,
MTA pulpotomy, and FC pulpotomy had higher quality evi­
dence at 24 months to support their use. MTA and FC 24-
month success rates were the highest of all pulpotomy types in
this time frame and did not differ statistically. At 24 months,
MTA, FC, and FS success rates were all significantly better in
their meta-analyses than CH ; therefore, we do not recom­
m end the use of C H pulpotom y. MTA was significantly
better than FS with a moderate level of evidence, which calls
into question using FS as a pulpotomy agent. At 18 months,
FC ’s pulpotom y success rate was significantly better than
NaOCl’s but with a low quality of evidence. Practitioners who
use NaOCl pulpotomy should consider other more successful
pulpotomy types with higher qualities of evidence. All other
pulpotom y agents’ success rates had shorter follow-ups and
either did not differ statistically from MTA and FC, or there
were no studies we could use in a meta-analysis comparison.
Results in context with previous studies
The current review was more complete than previous sys­
tematic reviews for evaluating the literature for treatment of
22 V ITA L PULP T HERAPY
deep caries in primary teeth. A recent Cochrane review, pub­
lished in 2014,19 on the treatment of deep caries in primary
teeth, had several limitations in comparison with the current
review. First, it did not include the treatment option of IPT
and limited reporting of treatment approaches for vital primary
teeth to DPC and pulpotomy. The omission of IPT was con­
sidered to be a shortcoming, as studies of IPT have been avail­
able in the literature for more than 15 years, and it is listed as
a treatment option in the Clinical Guidelines for Pulp Therapy
of Primary Teeth of the AAPD.3 The current review found
six IPT articles that we included in our systematic review in­
volving four investigations at different time frames.6' 10,28 IPT
had a 94.4 percent success rate at 24 months, which we believe
supports it’s use as a VPT option.
A second issue of the Cochrane review was their failure to
adequately represent the success rates of formocresol in their
final recommendations. Despite its equal performance with
MTA and FS, the authors omitted reference to it in their final
conclusions. The single sentence used to downgrade formocresol
appeared to misrepresent the reference, Milnes78, who actually
specified that formocresol is safe as currently used in dentistry
for vital pulp therapy/8 Milnes noted that while the Interna­
tional Agency for Research on Cancer (IARC) reclassified
formaldehyde in 2014 from a “probable” to a “known” carcino­
gen, this classification is not an assessment of risk but an
attempt to answer the question of whether, under any circum­
stances, a substance could produce cancer in humans. 78 Milnes
explained “the facts are that formaldehyde occurs naturally
throughout the body, there are multiple pathways for detoxifi­
cation, and only microgram quantities of formaldehyde are
applied to pulp tissues during pulpotomy procedures for mere
m inutes”. He further stated, “considering these facts, the
negative findings provide convincing evidence that exposure of
children to the formaldehyde component of formocresol during
a pulpotomy is insignificant and inconsequential.”78 While we
do not intend to promote formocresol as a primary tooth VPT
agent, we felt the evidence showed that FC ’s success rates
equaled or surpassed those of other studied agent, and this fact
must be acknowledged.
A third limitation of the Cochrane Review was missing
studies. The current review found five studies from 2012 and
before that were not included in the 2014 Cochrane Re­
view.6'39'47'79,80 In addition, the Cochrane review included data
from six-months observations, which is a very short time to
determine success of VPTs and a time frame when most treat­
ment options are found to be highly successful. Finally, the
Cochrane review did not attem pt to standardize the widely
divergent reports of success and failure rates as was done in
the present report.
Strengths and weaknesses
T he current systematic review followed the recom m enda­
tions in the Cochrane Handbook for Systematic Reviews of
Interventions.35 We conducted screening and data extraction in
duplicate, pooled split-mouth and parallel trials, and conducted
the assessment of the quality of the evidence using the GRADE
approach.
We believe an additional strength of this study was the
development of rules that allowed recalculation of the success
rates reported in the different studies to ensure they were all
standardized for use in the meta-analysis. However, the un­
standardized reporting, in some cases, made it impossible to
accurately combine data for the meta-analysis.

Weaknesses of the present report included the limited
responses that we received from primary study authors to clarify
issues related to risk of bias or data reporting. Another weak­
ness was that the only non-English language articles reviewed
were in Spanish and Portuguese. We also were unable to assess
publication bias due to the limited number of included studies
per outcome; however, many of our included studies reported
no differences between groups, suggesting limited publica­
tion bias.
We had to combine trials judged as unclear risk of bias
with low risk of bias in the analyses due to the relatively small
number of trials found in each comparison and the extremely
small number of low risk of bias trials. Since the unclear do­
main was related to insufficient reporting from the original
authors, it was not possible to estimate the amount and direc­
tion of bias, if any. It should be noted that all included trials
satisfied one of the critical domains from the Cochrane tool,
randomization, which reduces selection bias.
Quality of evidence/limitations
The quality of evidence for the three treatment options, IPT,
DPC, and pulpotomy, varied from high to very low. This vari­
ance was dependent on issues of precision associated with sample
size and the result of limited times of observation for some
agents/techniques. An assessment of the forest plots comparing
the different pulpotomy treatments at different times of obser­
vation, e.g. 12, 18, 24, months, demonstrated that differences
between treatm ent medicaments became larger with longer
times of observation. Some of our studies had only 12 or 18
months observation, and the differences in those outcomes
might have become significantly different with increased times
of observation. For instance the meta-analysis of FC vs. NaOCl
at 12 months showed the RR=1.06; P=0.22, but at 18 months
the meta-analysis showed the RR=1.20; P=0.01, indicating a
significant difference. Therefore our reported finding that
NaOCL was less successful than FC for pulpotomy may have
increased in magnitude with longer times of observation.
Implications for practice
These results suggest the amount of caries-affected dentin re­
moval is an important factor that should be considered when
choosing a VPT treatm ent option. Indirect pulp therapy
involves less tissue removal compared to DPC and pulpotomy.
IPT does not expose or damage the pulp and represents a
biological approach to pulp therapy whereby the pulp is
allowed to heal from a deep caries insult. Additionally, IPT
success rates appear to be equal or more successful compared
to pulpotomy and DPC success rates at 24 months. The DPC
meta-analysis result at 24 months indicated the use of DPC
for pinpoint carious exposures (one mm or smaller) may be a
viable treatment option. Pulpotomy continues to be a success­
ful option for VPT when carious exposures are encountered.
Our results suggest that two agents, MTA and FC, are the most
successful over time with the highest level of evidence.
The results of this systematic review and meta-analysis will
inform a revised Guideline for Pulp Therapy for Primary Teeth
for the AAPD that will now be evidence-based. Our findings
support the diagnostic decision that the indications are the
same for all three VPT treatm ent options investigated, IPT,
DPC and pulpotomy, and the only variable driving the treat­
ment approach is the amount of caries-affected dentin removed.
The comparable success rates for all three VPT treatment options
provide more latitude in treatment choices for the practitioner
P E D IA T R IC D E N T IS T R Y V 39 I NO 1 JA N I F EB 17
in managing a vital primary tooth with deep caries. A more
conservative approach to caries removal, that of leaving caries-
affected dentin to avoid pulp exposure (IPT), was supported
by the evidence reported here. When removal of caries resulted
in a small carious pulp exposure of one mm or less, our meta­
analysis showed that direct pulp capping, a procedure that the
current Practice Guideline does not recommend, is an option.
This review also suggested that the choice of IPT liner or DPC
capping agent does not impact success of these procedures.
These findings will need to be reflected in the revised Practice
Guideline. The choice of pulpotomy agents (MTA or FC) can
now be recommended with a higher degree of confidence.
Additionally some materials can be removed from the choices
(CH) or given reduced recommendations (NaOCl, laser, FS)
based on the evidence presented in our meta-analyses.
O ne o f our objectives was to determ ine if specific
moderators/factors affected VPT success. Our ability to accu­
rately do so was impacted by the small num ber o f trials in
each comparison, as well as incomplete and missing data and
unbalanced event distribution, all of which severely limited our
power to test the effect of many of the moderators-/factors. Only
the type of final restoration, study design, and use of rubber
dam could be evaluated, and there were no significant differ­
ences in the effect of any of these moderators/factors on the
success of the VPT procedures.
Implications for research
Authors of future studies are strongly encouraged to adopt a
standardized approach to report the results of their investiga­
tions, such as the CONSORT checklist.88 The current review
found serious problems during data extraction, which had the
potential to affect the validity and reliability of the data in­
cluded in the meta-analyses. The inconsistencies and lack of
standardization in the way success rates were reported made it
impossible to combine data from studies without impacting
the validity of the final analyses. For example, some authors re­
ported normal, uneventfully exfoliated teeth as dropouts and
removed them from the calculations for success.17,24'26 Others
reported a failure in one time frame, but did not carry that fail­
ure forward to the next time frame.17,24,25 Some studies reported
overall success as various combinations of clinical and radio-
graphic success.6,10,11,13,18,27'31 Some authors provided detailed
flow charts, but the data presented in these flow charts did not
match the data reported in the text of the articles. Others had
no charts or tables and only reported data in the text of the
manuscript. One well-designed, long-term study could not be
included, because it was not possible to extract the data due to
their unique method of reporting observations.89
In order to ensure the data were counted the same across
studies, we developed a set of rules to apply during data extrac­
tion to deal with the inconsistent methods of reporting results;
therefore, our data reported for some studies does not match
the data reported in those studies.6,10,12,16,17,21,22,24'29 To overcome
these issues of inconsistent reporting and the need to recalcu­
late data, it is recommended a set of guidelines for reporting
outcome data be developed. Standardization of dropout re­
porting of normally exfoliating teeth is needed, including
flowcharts that consistently show failures over time. It is
recommended that outcomes be reported as failed teeth divided
by total teeth for each time frame for each material, as well
as percentages. Additionally, criteria for determining clinical,
radiographic, and overall failure varied among studies, an issue
previously noted by Smail-Faugeron, et al.90
V IT A L P U L P T H E R A P Y 23
P E D IA TR IC D E N TIS T R Y V 3 9 I NO 1 JAN I FEB 17
Conclusions
Based on the results of this systematic review and the meta­
analyses of VPT in primary teeth, the following conclusions
can be made:
1. The success rate for IPT was 94.4 percent at 24
months, and the finding that liner material did not
affect the IPTsuccess was based on a moderate quality
of evidence.
2. The success rate for DPC of one mm or less carious
pulp exposure was 88.8 percent at 24 months, and
the finding that the pulp capping agents did not
affect success was based on a low quality of evidence.
3. The combined success rate for all pulpotomy agents
in the meta-analyses (MTA, FC, FS, N aO Cl, CH,
and laser) was 82.6 percent.
4. The highest pulpotom y success rates at 24 months
were FC (85.6 percent) vs. MTA (89.6 percent), and
they were not significantly (/?=0.15) different based
on a high quality of evidence.
5. Pulpotomy success rates at 24 months for MTA (89.0
percent), FC (79.0 percent), and FS (82.1 percent)
were all significantly better than CH (46 percent,
41.4 percent and 52.8 percent respectively), with a
moderate to low quality level of evidence. CH is not
recommended as a pulpotomy agent for primary teeth.
6. Pulpotomy success rate at 24 months for MTA (92.2
percent) was higher than FS (79.3 percent), approach­
ing significance (P=0.06), and was based on a mod­
erate quality of evidence.
7. Pulpotomy success rate at 18 months for NaOCl was
82.9 percent and was significantly less than FC (98.1
percent), based on a moderate quality of evidence.
8. Rules about the reporting of pulp therapy data con­
cerning exfoliations, dropouts, and success/failures
should be developed to standardize the extraction of
data for use in meta-analyses.
Acknowledgments
The authors wish to thank the American Academy of Pediatric
Dentistry, Chicago, 111., USA, for their financial and adminis­
trative support as well as the following for their assistance in
the developm ent of this systematic review: Dr. Peter Day,
International Association of Dental Traumatology; Ms. Mary
Foley, Medicaid SCHIP Dental Association; Dr. Anna Fuks,
Hadassah School of Dental Medicine, Jerusalem, Israel; Dr. Tom
Ison, Council on Dental Benefit Programs, AAPD; Dr. Janice
Jackson, Council on Postdoctoral Education, AAPD; Dr. Jessica
Lee, Evidence-Based Dentistry Committee, AAPD; Dr. Stephanie
Lepsky, National Association of Dental Plans; Dr. Randy Lout,
Council on Clinical Affairs, AAPD; Dr. Jan Mitchell, Academy
of General Dentistry; Dr. Lynn Mouden, Centers for Medicare
and Medicaid Services; Dr. Neva Penton-Eklund, Council on
Pre-Doctoral Education, AAPD; Dr. Anne Wilson, Council on
Scientific Affairs, AAPD; Dr. David W itherspoon, American
Association of Endodontists; and Dr. Timothy Wright, Amer­
ican Dental Association.
Authors’ contribution to study. Dr. James A. Coll: pri­
mary author of study concept and design, data extraction, ini­
tial writing, and critical revision of manuscript. Dr. N. Sue
Seale: study concept and design, data extraction, initial writing,
and critical revision of manuscript. Dr. Kaaren Vargas: study
24 V IT A L PULP TH ER APY
concept and design, data extraction, initial writing, and revision
of manuscript. Dr. Abdullah A. Marghalani: statistical analysis
and critical revision of manuscript. Dr. Shahad AlShamali: cri­
tical revision of included studies and statistical support. Ms.
Laurel Graham: manager of systematic review process, initial
writing and critical revision of manuscript, and search strategy
development.
Conflicts o f interest. Dr. Seale is Editor in Chief of the
journals of the American Academy of Pediatric Dentistry, and
receives payment for her services. Dr. James Coll is a Section
Editor for Pediatric Dentistry, and Drs. Marghalani, and Vargas
are Ad Hoc reviewers for Pediatric Dentistry.
References
1. Dye BA, Li X, Thornton-Evans G. Oral Health Dispari­
ties as Determined by Selected Healthy People 2020 Oral
Health Objectives for the United States, 2009-2010. US
Department of Health and Human Services, Centers for
Disease Control and Prevention, N ational Center for
Health Statistics; 2012.
2. Fuks A, Kupietzki A, Guelmann M. Pulp therapy for the
primary dentition. In: Pediatric Dentistry: Infancy through
Adolescence. 5e ed. St. Louis, Mo.: Elsevier/Saunders;
2013:333.
3. American Academy of Pediatric Dentistry Reference Ma­
nual. 2016 Guideline on Pulp Therapy for Primary and
Immature Permanent Teeth. Pediatr Dent 2016;38(6):
280-8.
4. Tomes J, Tomes SC. A System of Dental Surgery. London:
J. & A. Churchill; 1873:336.
5. Hunter FA. Saving pulps. A queer process. Items Interest.
1883;5:352.
6. Buyukgural B, Cehreli ZC. Effect of different adhesive
protocols vs. calcium hydroxide on primary tooth pulp
with different remaining dentin thicknesses: 24-month
results. Clin Oral Investig 2008; 12(1 ):91-6.
7. Casagrande L, Bento LW, Rerin SO, Lucas Ede R, Dalpian
DM , de Araujo FB. In vivo outcomes of indirect pulp
treatment using a self-etching primer versus calcium hy­
droxide over the demineralized dentin in primary molars.
J Clin Pediatr Dent 2008;33(2): 131-5.
8. Casagrande L, Falster CA, Hipolito V, et al. Effect of adhesive
restorations over incomplete dentin caries removal: 5-year
follow-up study in primary teeth. J Dent Child Chic 111
2009;76(2): 117-22.
9. Falster CA, Araujo FB, Straffon LH, Nor JE. Indirect pulp
treatment: in vivo outcomes of an adhesive resin system vs.
calcium hydroxide for protection of the dentin-pulp com­
plex. Pediatr Dent 2002;24(3):241-8.
10. Marchi JJ, Araujo FB, Froner AM, Straffon LH, Nor JE.
Indirect pulp capping in the primary dentition: a 4 year
follow-up study. J Clin Pediatr Dent 2006;31(2):68-71.
11. Demir T, Cehreli ZC. Clinical and radiographic evalua­
tion of adhesive pulp capping in primary molars follow­
ing hemostasis with 1.25% sodium hypochlorite: 2-year
results. Am J Dent 2007;20(3):182-8.
12. Fallahinejad Ghajari M, Asgharian Jeddi T, Iri S, Asgary
S. Treatm ent outcomes o f primary molars direct pulp
capping after 20 months: a randomized controlled trial.
Iran Endod J 2013;8(4): 149-52.

13. Ulusoy AT, Bayrak B, Bodrumlu EH. Clinical and radio­
logical evaluation of calcium sulfate as direct pulp cap­
ping material in primary teeth. Eur J Paediatr Dent 2014;
15(2): 127-31.
14. Sweet CA. Treatment and Maintenance of Pulpless Deci­
duous Teeth. **Read before the Section on Mouth Hygiene
and Preventive D entistry at the Seventy-Sixth Annual
Session of the American D ental Association, St. Paul,
M inn., August 8, 1935. J Am Dent Assoc 1935;22(11):
1972-5.
15. Ni Chaollai AN, Monteiro J, Duggal MS. The teaching
of management of the pulp in primary molars in Europe:
a preliminary investigation in Ireland and the UK. Eur
Arch Paediatr Dent 2009;10(2):98-103.
16. Sushynski JM, Zealand CM, Botero TM, et al. Compari­
son of gray mineral trioxide aggregate and diluted formo-
cresol in pulpotomized primary molars: a 6- to 24-month
observation. Pediatr Dent 2012;34(5): 120-8.
17. Huth KC, Hajek-Al-Khatar N, Wolf P, Ilie N, Hickel R,
Paschos E. Long-term effectiveness of four pulpotomy
techniques: 3-year randomised controlled trial. Clin Oral
Investig 2012; 16(4): 1243-50.
18. Trairatvorakul C, Sastararuji T. Indirect pulp treatment vs.
antibiotic sterilization of deep caries in mandibular pri­
mary molars. Int J Paediatr Dent 2014;24(1):23-31.
19. Smail-Faugeron V, Courson F, Durieux P, Muller-Bolla M,
Glenny A-M, Fron Chabouis H. Pulp treatment for ex­
tensive decay in primary teeth. Cochrane Database Syst
Rev 2014;8:CD003220.
20. Marghalani AA, Omar S, Chen J-W. Clinical and radio-
graphic success of mineral trioxide aggregate compared
with formocresol as a pulpotomy treatm ent in primary
molars: A systematic review and meta-analysis. J Am Dent
Assoc 2014;145(7):714-21.
21. Peng L, Ye L, Tan H, Zhou X. Evaluation of the formo­
cresol versus mineral trioxide aggregate primary molar
pulpotomy: a meta-analysis. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 2006;102(6):e40-4.
22. Coll j, Seale NS, Vargas K, Chi DL, Marghalani AA, Gra­
ham L. Protocol for a Systematic Review and Meta-analysis
of Vital Pulp Therapy for Children with Deep Caries in
the Primary Dentition. Pediatr Dent 2015;37(5):4l 8-21.
23. Hausner E, Waffenschmidt S, Kaiser T, Simon M. Routine
development of objectively derived search strategies. Syst
Rev 2012; 1:19.
24. Fernandes AP, Lourenco Neto N, Teixeira Marques NC, et
al. Clinical and radiographic outcomes of the use of Low-
Level Laser Therapy in vital pulp of primary teeth. Int J
Paediatr Dent 2015;25(2): 144-50. Epub 2014.
25. Waterhouse PJ, N unn JH , W hitworth JM. An investiga­
tion of the relative efficacy of Buckley’s Formocresol and
calcium hydroxide in primary molar vital pulp therapy.
Br Dent J 2000; 188( l):32-6.
26. Agamy HA, Bakry NS, Mounir MM, Avery DR. Comparison
of mineral trioxide aggregate and formocresol as pulp­
capping agents in pulpotomized primary teeth. Pediatr
Dent 2004;26(4):302-9.
27. Trairatvorakul C, D etsom boonrat P. Success rates of a
mixture of ciprofloxacin, metronidazole, and minocycline
antibiotics used in the non-instrumentation endodontic
treatm ent of m andibular prim ary molars with carious
pulpal involvement. Int J Paediatr D ent 2 012;22(3):
217-27.
P EDIATRIC D E N TIS T R Y V 39 / NO 1 JAN / FEB 17
28. Casagrande L, Bento LW, Dalpian DM, Garcia-Godoy F,
de Araujo FB. Indirect pulp treatment in primary teeth:
4-year results. Am J Dent 2010;23(l):34-8.
29. Tuna D, Olmez A. Clinical long-term evaluation of MTA
as a direct pulp capping material in primary teeth. Int
Endod J 2008;4l(4):273-8.
30. Noorollahian H. Comparison of mineral trioxide aggregate
and formocresol as pulp medicaments for pulpotomies in
primary molars. Br Dent J 2008;204(11):E20. Epub 2008
April 18.
31. Aminabadi NA, Farahani RM, Oskouei SG. Formocresol
versus calcium hydroxide direct pulp capping of human
primary molars: two year follow-up. J Clin Pediatr Dent
2010;34(4):317-21.
32. Holan G, Eidelman E, Fuks AB. Long-term evaluation of
pulpotomy in primary molars using mineral trioxide aggre­
gate or formocresol. Pediatr Dent 2005;27(2): 129-36.
33. Zurn D, Seale NS. Light-cured calcium hydroxide vs. for­
mocresol in human primary molar pulpotomies: a ran­
domized controlled trial. Pediatr Dent 2008;30( 1):34-41.
34. Roberts JD. Indirect pulp treatment vs. Formocresol Pul­
potomy in Human Primary Molars: A Randomized Con­
trolled Trial. Unpublished manuscript. 2008 Submitted
to The Office of Research and Graduate Studies of The
Texas A&M University System Health Science Center
in 2009. Available at: “http://pqdtopen.proquest.com /
pubnum/l462151.html”. Accessed February 7, 2017.
35. Higgins J, Green S (editors). Cochrane Handbook for Sys­
tematic Reviews of Interventions Version 5.1.0.; 2011.
Available at: “www.handbook.cochrane.org”. Accessed
September 22, 2016.
36. Grading of Recommendations Assessment, Development
and Evaluation (short GRADE) working group. Available
at: “http://www.gradeworkinggroup.Org/.GRADE”, “http:
/ /www. gradewo rki nggro up.org/ ”.
37. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measur­
ing inconsistency in meta-analyses. BMJ 2003;327(74l4):
57-560.
38. Cantekin K, Giimii? H. Success Rates of Ankaferd Blood
Stopper and Ferric Sulfate as Pulpotomy Agents in Pri­
mary Molars. Int Sch Res Notices 2014;2014:819065.
eCollection 2014.
39. Casas MJ, Layug MA, Kenny DJ, Johnston DH, Judd PL.
Two-year outcomes of primary molar ferric sulfate pulpo­
tomy and root canal therapy. Pediatr Dent 2003;25(2):
97-102.
40. Casas MJ, Kenny DJ, Johnston DH, Judd PL. Long-term
outcomes of primary molar ferric sulfate pulpotomy and
root canal therapy. Pediatr Dent 2004;26(l):44-8.
41. Howley B, Seale NS, McWhorter AG, Kerins C, Boozer
KB, Lindsey D. Pulpotomy versus pulpectomy for carious
vital primary incisors: randomized controlled trial. Pediatr
Dent 2012;34(5): 112-9.
42. Malekafzali B, Shekarchi F, Asgary S. Treatment outcomes
of pulpotomy in primary molars using two endodontic
biomaterials. A 2-year randomised clinical trial. Eur J
Paediatr Dent 2011; 12(3): 189-93.
43. Odabas ME, Cinar C, Tulunoglu O, Isik B. A new haemo­
static agent’s effect on the success of calcium hydroxide
pulpotomy in primary molars. Pediatr Dent 2011;33(7):
529-34.
V ITA L PULP T H E R A P Y 25
PED IA TR IC D E N T IS TR Y V 3 9 I NO 1 JAN / FEB 17
44. Sakai VT, Moretti ABS, Oliveira TM, et al. Pulpotomy of
human primary molars with MTA and Portland cement:
a randomised controlled trial. Br Dent J 2009;207(3):E5;
discussion 128-9.
45. Yaman E, Gorken F, Pinar Erdem A, Sepet E, Aytepe Z.
Effects of folk medicinal plant extract Ankaferd Blood
Stopper (®) in vital primary molar pulpotomy. Eur Arch
Paediatr Dent 2012; 13 (4): 197-202.
46. Uloopi KS, Vinay C, Ratnaditya A, Gopal AS, Mrudula
KJN, Rao RC. Clinical Evaluation of Low Level Diode
Laser Application For Primary Teeth Pulpotomy. J Clin
Diagn Res 2016;10(l):ZC67-70.
47. Nematollahi H, Sahebnasagh M, Parisay I. Comparison
of electrosurgical pulpotomy with zinc oxide eugenol or
zinc polycarboxylate cements sub-base. J Clin Pediatr Dent
2011;36(2):133-7.
48. Aminabadi NA, Farahani RM, Gajan EB. A clinical study
ol formocresol pulpotomy versus root canal therapy of
vital primary incisors. J Clin Pediatr Dent 2008;32(3):
211-4.
49. Aminabadi NA, Huang B, Samiei M, Agheli S, Jamali Z,
Shirazi S. A Randomized Trial Using 3Mixtatin Compared
to MTA in Primary Molars with Inflammatory Root Re­
sorption: A Novel Endodontic Biomaterial. J Clin Pediatr
Dent 2016;40(2):95-102.
50. Ibricevic H, Al-Jame Q. Ferric sulphate and formocresol
in pulpotomy of primary molars: long term follow-up
study. Eur ] Paediatr Dent 2003;4:28-32.
51. Khorakian F, Mazhari F, Asgary S, et al. Two-year out­
comes of electrosurgery and calcium-enriched mixture
pulpotomy in primary teeth: a randomised clinical trial.
Eur Arch Paediatr Dent 2014; 15(4):223-8.
52. Saltzman B, Sigal M, Clokie C, Rukavina J, Titley K,
Kulkarni GV. Assessment of a novel alternative to conven­
tional formocresol-zinc oxide eugenol pulpotomy for the
treatment of pulpally involved human primary teeth: diode
laser-mineral trioxide aggregate pulpotomy. Int J Paediatr
Dent 2005;15(6):437-47.
53. Shumayrikh NM, Adenubi JO. Clinical evaluation of
glutaraldehyde with calcium hydroxide and glutaraldehyde
with zinc oxide eugenol in pulpotomy of primary molars.
Endod Denr Traumatol 1999;15(6):259-64.
54. Farsi N, Alamoudi N, Balto K, Mushayt A. Success of
mineral trioxide aggregate in pulpotomized primary mo­
lars. J Clin Pediatr Dent 2005;29(4):307-l 1.
55. Ibricevic H, Al-Jame Q. Ferric sulfate as pulpotomy agent
in primary teeth: twenty month clinical follow-up. J Clin
Pediatr Dent 2000;24(4):269-72.
56. Jabbarifar SE, Khademi AA, Ghasemi D. Success rate of
formocresol pulpotomy versus mineral trioxide aggregate
in human primary molar tooth. J Res Med Sciences 2004;
6:304-307. Available at: “www.SID.ir”. Accessed Decem­
ber 13, 2016.
57. Jayam C, Mitra M, Mishra J, Bhattacharya B, Jana B. Evalu­
ation and comparison of white mineral trioxide aggregate
and formocresol medicaments in primary tooth pulpo­
tomy: Clinical and radiographic study. J Indian Soc Pedod
Prev Dent 20l4;32(l):13-8.
58. Markovic D, Zivojinovic V, Vucetic M. Evaluation of three
pulpotomy medicaments in primary teeth. Eur J Paediatr
Dent 2005;6(3): 133-8.
26 V ITAL PULP TH ER APY
59. Olatosi OO, Sote EO, Orenuga OO. Effect of mineral tri­
oxide aggregate and formocresol pulpotomy on vital pri­
mary teeth: a clinical and radiographic study. Niger J Clin
Pract 2015;18(2):292-6.
60. Ruby JD, Cox CF, Mitchell SC, Makhija S, Chompu-Inwai
P, Jackson J. A randomized study of sodium hypochlorite
versus formocresol pulpotomy in primary molar teeth.
Int J Paediatr Dent 2013;23(2): 145-52.
61. Shabzendedar M, Mazhari F, Alami M, Talebi M. Sodium
hypochlorite vs. formocresol as pulpotomy medicaments
in primary molars: 1-year follow-up. Pediatr Dent 2013;
35(4):329-32.
62. Vargas KG, Packham B, Lowman D. Preliminary evalua­
tion of sodium hypochlorite for pulpotomies in primary
molars. Pediatr Dent 2006;28(6):511-7.
63. Farsi DJ, El-Khodary HM, Farsi NM, El Ashiry EA,
Yagmoor MA, Alzain SM. Sodium Hypochlorite Versus
Formocresol and Ferric Sulfate Pulpotomies in Primary
Molars: 18-month Follow-up. Pediatr Dent 2015:37(7):
535-40.
64. Rajasekharan S, Martens L, Vandenbulcke J, Jacquet W,
Bottenberg P, Cauwels R. Efficacy of three different pul­
potomy agents in primary molars - A randomised control
trial. Int EndodJ 2016. doi: 10.1111/iej. 12619.
65. Fernandez CC, Martinez SS, Jimeno FG, Lorente Rodriguez
Al, Mercade M. Clinical and radiographic outcomes of
the use ol lour dressing materials in pulpotomized primary
molars: a randomized clinical trial with 2-year follow­
up. Int J Paediatr Dent 2013;23(6):400-7.
66. Erdem AP, Guven Y, Balli B, et al. Success rates of mineral
trioxide aggregate, ferric sulfate, and formocresol pulpoto­
mies: a 24-month study. Pediatr Dent 2011 ;33(2): 165-70.
67. Moretti AB, Sakai VT, Oliveira TM, et al. The effective­
ness of mineral trioxide aggregate, calcium hydroxide and
formocresol for pulpotomies in primary teeth. Int Endod
J 2008;4l(7):547-55.
68. Sonmez D, Sari S, Cetinba T. A Comparison of four pul­
potomy techniques in primary molars: a long-term follow­
up. J Endod 2008;34(8):950-5.
69. Subramaniam P, Konde S, Mathew S, Sugnani S. Mineral
trioxide aggregate as pulp capping agent for primary teeth
pulpotomy: 2 year follow up study. J Clin Pediatr Dent
2009;33(4):311-4.
70. Yildirim C, Basak F, Akgun OM, Polat GG, Altun C. Clin­
ical and Radiographic Evaluation of the Effectiveness of
Formocresol, Mineral Trioxide Aggregate, Portland Ce­
ment, and Enamel Matrix Derivative in Primary Teeth
Pulpotomies: A Two Year Follow-Up. J Clin Pediatr Dent
2016;40(1): 14-20.
71. Akcay M, Sari S. The effect of sodium hypochlorite appli­
cation on the success of calcium hydroxide and mineral
trioxide aggregate pulpotomies in primary teeth. Pediatr
Dent 2014;36(4):316-21.
72. Al-Mutairi MA, Bawazir OA. Sodium hypochlorite versus
Formocresol in primary molars pulpotomies: a random­
ized clinical trial. Eur J Paediatr Dent 2013; 14( 1):33-6.
73. Eidelman E, Holan G, Fuks AB. Mineral trioxide aggre­
gate vs. formocresol in pulpotomized primary molars: a
preliminary report. Pediatr Dent 2001;23(1): 15-8.
74. Celik B, Ataf AS, Cehreli ZC, Uysal S. A randomized trial
of mineral trioxide aggregate cements in primary tooth
pulpotomies. J Dent Child (Chic 111) 2013;80(3): 126-32.

75. Doyle TL, Casas MJ, Kenny DJ, Judd PL. Mineral trioxide
aggregate produces superior outcomes in vital primary
molar pulpotomy. Pediatr Dent 2010;32(l):4l-7.
76. Durmus B, Tanboga I. In vivo evaluation of the treatment
outcom e of pulpotom y in primary molars using diode
laser, formocresol, and ferric sulphate. Photomed Laser
Surg 20l4;32(5):289-95.
77. Cuadros-Fernandez C, Lorente Rodriguez AI, Saez-Martinez
S, et al. Short-term treatment outcome of pulpotomies
in primary molars using mineral trioxide aggregate and
Biodentine: a randomized clinical trial. Clin Oral Investig
20l6;20(7):l639-45.
78. Milnes AR. Is formocresol obsolete? A fresh look at the
evidence concerning safety issues. Pediatr Dent 2008;30
(3):237-46.
79. Liu JF. Effects of Nd:YAG laser pulpotom y on human
primary molars. J Endod 2006;32(5):404-7.
80. Odabas ME, Alacam A, Sillelioglu H, Deveci C. Clinical
and radiographic success rates of mineral trioxide aggre­
gate and ferric sulphate pulpotomies performed by dental
students. EurJ Paediatr Dent 2012; 13(2): 118-22.
81. Alacam A. Long term effects of primary teeth pulpoto­
mies with formocresol, glutaraldehyde-calcium hydroxide
and glutaraldehyde-zinc oxide eugenol on succudaneous
teeth. J Pedod 1989; 13(4):307-13.
82. Fei AL, Udin RD, Johnson R. A clinical study of ferric sul­
fate as a pulpotomy agent in primary teeth. Pediatr Dent
1991;13(6):327-32.
83. Havale R, Anegundi RT, Indushekar K, Sudha P. Clinical
and radiographic evaluation of pulpotomies in primary
molars with formocresol, glutaraldehyde and ferric sul­
phate. Oral Health Dent Manag 2013; 12(l):24-31.
P E D IA T R IC D E N T IS T R Y V 39 I NO 1 JA N I FEB 17
84. Musale PK, Soni AS. Clinical Pulpotomy Trial of Copaifera
Langsdorffii Oil Resin Versus Formocresol and W hite
Mineral Trioxide Aggregate in Primary Teeth. Pediatr Dent
2016;38(2):5-12.
85. Srinivasan D, Jayanthi M. Comparative evaluation of for­
mocresol and mineral trioxide aggregate as pulpotomy
agents in deciduous teeth. Indian J Dent Res 2011;22(3):
385-90.
86. Fuks AB, Holan G, Davis JM, Eidelman E. Ferric sulfate
versus dilute formocresol in pulpotomized primary mo­
lars: long-term follow up. Pediatr D ent 1997; 19(5):
327-30.
87. Al-Mutairi MA, Bawazir OA. Sodium hypochlorite versus
Formocresol in primary molars pulpotomies: a randomized
clinical trial. Eur J Paediatr Dent Off J Eur Acad Paediatr
Dent 2013;l4(l):33-6.
88. Schulz KF, Altm an DG, M oher D, C onsort Group.
CONSORT 2010 statement: updated guidelines for re­
porting parallel group randomised trials. Int J Surg 2011;
9(8):672-7.
89. Mettlach SE, Zealand CM, Botero TM, et al.. Comparison
of mineral trioxide aggregate and diluted formocresol in
pulpotomized human primary molars: 42-month follow­
up and survival analysis. Pediatr D ent 2013;35(3):
E87-94.
90. Smail-Faugeron V, Fron Chabouis H, Durieux P, Attal J-P,
Muller-Bolla M, Courson F. Development of a core set of
outcomes for randomized controlled trials with multiple
outcomes—example of pulp treatments of primary teeth for
extensive decay in children. PloS One 2013;8(l):e51908.
V IT A L P U L P T H E R A P Y 27
Copyright of Pediatric Dentistry is the property of American Society of Dentistry for Children
and its content may not be copied or emailed to multiple sites or posted to a listserv without
the copyright holder's express written permission. However, users may print, download, or
email articles for individual use.