A comparison of calcium hydroxide/iodoform paste and zinc oxide eugenol as root filling materials for pulpectomy in primary teeth: A systematic review and meta‐analysis

Abstract Zinc oxide eugenol (ZOE) has traditionally been used as a root filling material in primary teeth pulpectomy. Calcium hydroxide and iodoform (Ca(OH)2/iodoform) may have advantages as a root canal filling material to evaluate treatment success of Ca(OH)2/iodoform pulpectomy in primary teeth compared with ZOE based on clinical and radiographical criteria. All human clinical studies reporting clinical and radiographical outcomes of Ca(OH)2/iodoform compared with ZOE in primary teeth pulpectomy were identified in digital bibliographic databases. Two authors independently selected studies and extracted relevant study characteristics. Success of treatment was based on an accomplishment of specific clinical and radiographical criteria. Meta‐analyses were performed to appraise study heterogeneity and aggregated statistics. Out of 5,000 articles identified in initial search, 15 articles met all inclusion criteria, while 10 were included in the meta‐analyses. At 6‐ and 12‐month follow‐up, there were no statistically significant differences in the clinical and radiographical success rates of Ca(OH)2/iodoform and ZOE. However, ZOE was shown to have statistically significant higher success rates at ≥18‐month follow‐up. On the basis of the findings of this systematic review, we recommend that Ca(OH)2/iodoform be utilized for pulpectomy in primary teeth nearing exfoliation; conversely, ZOE should be utilized when exfoliation is not expected to occur soon. Future randomized control clinical trials with a long‐term follow‐up are needed before a reliable conclusion can be drawn as to the best pulpectomy material. The success of pulpectomy in primary teeth depends on selecting the ideal root canal filling material. It is challenging to select the appropriate filling materials for primary teeth. ZOE or ZOE/iodoform combined with Ca(OH)2 appears to be the materials of choice if primary teeth are not nearing exfoliation. More high‐quality randomized control clinical trials with a long‐term follow‐up period are needed before a reliable conclusion can be drawn as to the best pulpectomy material in primary teeth (systematic review registration number: CRD42016037563).

The criteria of an ideal root canal filling material in primary teeth are as follows: being antibacterial, resorbs at the same rate as the roots and not causing harms to the periapical area, and the development of the succedaneous tooth. Also, it should fill the canal easily, adhere to the wall of the canal, resorb if extruded beyond the apex, show radio-opaque appearance in the radiograph, and do not cause discoloration to the tooth (Garcia-Godoy, 1987;Rifkin, 1980). Zinc oxide eugenol (ZOE) has been the conventional root canal filling material used for primary teeth pulpectomy since 1930. ZOE has several disadvantages: low resorption rate (Erausquin & Muruzabal, 1967), causing irritation to the periapical area (Spedding, 1985), necrosis to bone and cementum (Hendry, Jeansonne, Dummett, & Burrell, 1982), and deflection of the permanent tooth bud (Coll & Sadrian, 1996). Studies report that the success rate of ZOE alone or with fixative medications as formocresol or iodoform ranges from 65% to 86% (Coll, Josell, & Casper, 1985;Holan & Fuks, 1993).
In 1920, calcium hydroxide (Ca(OH) 2 ), a silicone oil-based paste, was introduced by Hermann and has been widely used. Iodoform has been added to Ca(OH) 2 due to its antibacterial effect (Estrela, Estrela, Hollanda, Decurcio, & Pécora, 2006), healing properties, and ability to be resorbed when in excess (Nurko, Ranly, Garcia-Godoy, & Lakshmyya, 2000). The reported success rate for the combined Ca(OH) 2 /iodoform paste ranges from 84% to 100% (Reddy & Fernandes, 1996). Additional benefits of iodoform include its radiopacity, the ease with which it can be introduced and removed from the canal, negative effect on the succedaneous tooth, and its ability to be resorbed within 8 weeks once it has been extruded beyond the apex (Nurko & Garcia-Godoy, 1999).
The main disadvantage of Ca(OH) 2 /iodoform paste is a potential risk of intracanal resorption (Nurko et al., 2000).
There are no comprehensive studies that examine the clinical and radiographical outcomes of Ca(OH) 2 /iodoform as a pulpectomy material in primary teeth. Therefore, the aim of this systematic review (SR) and meta-analysis was to evaluate treatment success of Ca(OH) 2 / iodoform pulpectomy in primary teeth compared with ZOE based on clinical and radiographical criteria.

| MATERIALS AND METHODS
This study was written according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis statement (Liberati et al., 2009). It was registered in the International Prospective Register of Systematic Reviews with registration number CRD42016037563.

| Selection criteria
Studies reporting clinical and radiographical outcomes of Ca(OH) 2 / iodoform compared with ZOE pulpectomy in primary teeth were considered as eligible. The inclusion criteria were randomized and nonrandomized clinical trials comparing the clinical and/or radiographical outcomes of Ca(OH) 2 /iodoform versus ZOE pulpectomy in primary teeth of healthy children. The exclusion criteria were as follows: cross-sectional, retrospective, laboratory, and animal studies. We also excluded all studies investigating pulpectomy in permanent teeth, traumatic teeth, or primary teeth without a succedaneous tooth. Our last exclusion criterion was any research whose study population included special needs patients.

| Search strategy and data extraction
Search strategies were designed to identify all studies discussing the clinical and radiographical outcomes of Ca(OH) 2 /iodoform compared with ZOE used in primary teeth pulpectomy. Two commercial formulations of Ca(OH) 2 /iodoform prevail on the market, and these are Metapex (Meta Biomed Co. Ltd, Seoul, South Korea) and Vitapex (Neo Dental Chemical Products Co. Ltd, Tokyo, Japan;Nurko et al., 2000;Stuart, Schwartz, Beeson, & Owatz, 2006). Hence, the following set of keywords were used during the search: (calcium hydroxide OR Vitapex OR Metapex) AND (pulpectomy OR pulpectomies OR pulpectomized OR root canal treatment OR root canal filling) AND (primary teeth OR primary dentition OR deciduous teeth). We initially limited our search to articles published between 2003 and 2017 without restrictions on publication year or language. This search strategy yielded a total of 5,000 articles from three search engines, PubMed/MEDLINE (261), Google Scholar (3,850), and Scopus (89).
Our initial search was conducted in April 2016. A subsequent search that was performed in January 2018 revealed one additional study (Chen, Liu, & Zhong, 2017) for inclusion.
The titles of all studies were reviewed by two authors independently (R. S. N. and H. J. S.). Duplicate studies were excluded. After titles selection, the abstracts were reviewed. Studies were excluded when it was obvious that the paper was not discussing any clinical and the radiographical outcomes of Ca(OH) 2 /iodoform compared with ZOE in primary teeth pulpectomy. The selected studies were downloaded as full text papers and then screened in details by the same reviewers to confirm whether they fulfilled the inclusion criteria.
Cohen's κ statistic was done with value of 0.92 and 96.29% of agreement. Disagreement was settled by the third evaluator (A. A. E.).
Using a data extraction sheet, the reviewers next independently collected data from the selected studies. Variables included publication details (author and year), study setting, research methodology (study design, number and age of children, number of teeth, type of teeth, presence of a ZOE subgroup, ZOE, ZOE/iodoform or ZOE/iodoform combined with Ca(OH) 2 , and sample size in each group), follow-up period(s), and clinical and radiographical outcomes.
κ statistic was done with value of 0.82 and 98.13% of agreement.
Cases of disagreement were discussed between the evaluator until agreement was reached.
In this SR, we defined (treatment) success based on the accomplishment of specific clinical and radiographical criteria. The clinical criteria are as follows: no pain, no swelling, no abscess, no pain on percussion, and/or decreased in mobility. The radiographical criteria are a decrease or an absence of radiolucency when comparing postoperative imaging with X-rays taken preoperatively. No change in radiolucency was considered as an indicator of success in three clinical success (Chen & Liu, 2005;Gupta & Das, 2011;Subramaniam & Gilhotra, 2011). Hence, this criterion was also adopted as a measure of success in four clinical studies (Al-Ostwani, Al-Monaqel, & Al-Tinawi, 2016;Pramila, Muthu, Deepa, Farzan, & Rodrigues, 2016;Trairatvorakul & Chunlasikaiwan, 2008;Xiao-Fang & Xue-Bin, 2003).

| Quality appraisal
The quality of the methodology and results of the included studies were assessed using a modified version of the Consolidated Standards of Reporting Trials (CONSORT) 2010 checklist for clinical trials quality assessment (Schulz, Altman, & Moher, 2010).
The methods and results part of CONSORT consist of 15 categories with 25 items. We added two more items, that is, the number of operators performing the pulpectomies and in studies with multiple operators, inter-operator reliability with respect to intervention methodology and outcome measures assessed. One point was assigned per item; therefore, the scale ranged from a minimum of 0 to a maximum of 27. The reviewers then independently categorized studies according to the following scores: 19-27 indicated a low risk of bias (highquality study), 10-18 indicated a moderate risk of bias (moderatequality study), and 0-9 indicated a high risk of bias (low-quality study).
When there were discrepancies in categorization, reviewers discussed manuscript scoring until an agreement was reached. Although studies were not excluded for high bias risk, the categorizations were used for sensitivity analysis in the meta-analysis.
The quality of each study was ranked by two independent evaluators (R. S. N. and H. J. S.). Cases of disagreement were discussed between the evaluators until agreement was reached. No exclusion based on the risk of bias was done. Studies were then classified into high, moderate, and low quality for sensitivity analysis in the metaanalysis.

| Statistical analysis
Studies reporting the clinical and radiographical success rates of Ca(OH) 2 /iodoform paste compared with ZOE were incorporated in the meta-analyses using Review Manager 5.3 (The Nordic Cochrane Centre; ReviewManager [RevMan], 2014). The Mantel-Haenszel method was used to calculate a weighted average of odds ratios (ORs) and generate 95% confidence intervals (95% CIs; Landis, Sharp, Kuritz, & Koch, 2005) for the success rates of pulpectomy with Ca(OH) 2 /iodoform paste compared with ZOE across all studies. To determine whether the results of separate studies could be combined meaningfully, a statistical test of homogeneity was carried out. An inconsistency coefficient (I 2 ) was calculated taking into account Cochrane's heterogeneity statistic and the degrees of freedom for the sample size included in our meta-analysis. I 2 describes the level of heterogeneity within a sample that contributes to variation as opposed to chance. The value of >25%, 50%, and 75% represent low, moderate, and high heterogeneity, respectively (Higgins, Thompson, Deeks, & Altman, 2003).
ORs were pooled with fixed effect if no heterogeneity was identified in the meta-analysis and with random effect in case of heterogeneous studies (DerSimonian & Laird, 1986). The level of significance was set at <0.05. Z test was used to compare the clinical and radiographical success rates of Ca(OH) 2 /iodoform to ZOE in all follow-up periods in high-and moderate-quality studies. Success rates of high-quality studies were compared with success rates of moderate-quality studies using a chi-squared test. A funnel plot was used to visually represent heterogeneity within publications; Egger's test was used for quantitative analysis of heterogeneity (Egger, Davey Why this paper is important: • The success of pulpectomy in primary teeth depends on selecting the ideal root canal filling material. It is challenging to select the appropriate filling materials for primary teeth. • Zinc oxide eugenol or zinc oxide eugenol/iodoform combined with Ca(OH) 2 appears to be the materials of choice if primary teeth are not nearing exfoliation.
• More high-quality randomized control clinical trials with a long-term follow-up period are needed before a reliable conclusion can be drawn as to the best pulpectomy material in primary teeth. Smith, Schneider, & Minder, 1997). These analyses were performed using the Comprehensive Meta-Analysis program version 3.3.070.

| Sensitivity analysis
Meta-analysis is confounded by many factors; these factors are thought to be a possible cause of heterogeneity if present. Subgroup analyses were used to assess the stability of the results. Analysis were carried out on the basis of the clinical and radiographical success rates to evaluate the effect of type of intracanal irrigation, type of teeth, and the quality of the studies to investigate the source of heterogeneity.

| Level of evidence
For our SR, we developed both an evidence statement and clinical recommendations using a modification to the guidelines provided by Shekelle, Woolf, Eccles, and Grimshaw (1999). Clinical recommendations were classified on the basis of the strength of evidence by which they were supported, as determined by adherence to measurable components defined in our evidence statement. It is important to note that the classification of recommendations reflects the quality of scientific evidence supporting a given recommendation rather than its clinical importance using a system modified from that of Shekelle et al. (1999).

| Study selection
The searches yielded 5,000 potentially related titles ( Figure 1). After removing the duplicate studies (602 studies) and those not eligible after reviewing the abstract, the full text of 27 studies was retrieved and compared with the inclusion criteria.
We excluded 11 studies as follows: three studies without comparison group, three were review, one study was not compared with ZOE, and four studies reported Ca(OH) 2 without iodoform.
The total number of 16 studies were included in this SR ( Figure 1).
The included studies had different eligibility criteria as well as different study methodologies. Variations were present in the number of treatment visits, the latency to follow-up examination, the type of irrigation solution used, and the final restorative material used (Table 1). and ZOE/iodoform in one study (Pramila et al., 2016), and ZOE/iodoform and ZOE/iodoform combined with Ca(OH) 2 in one study (Ramar & Mungara, 2010
Our meta-analysis also investigated the effect of confounding factors on the clinical and radiographical success rates of Ca(OH) 2 compared with ZOE and ZOE/iodoform combined with Ca(OH) 2 . In the subgroup analysis, we excluded one study that reported the success rates of ZOE/iodoform because there was no sufficient data for comparison (Pramila et al., 2016). Possible confounders included intracanal irrigation, type of molars, and study quality.
Because studies reported the use of different intracanal irrigation materials, they were further subdivided into two groups: those in which sodium hypochlorite (NaOCl) was used and studies in which any other intracanal irrigation was used. We compared the effect of varying intracanal irrigation solutions only to the ZOE group,

FIGURE 3
Forest plot for meta-analysis of the clinical and radiographical success rates of Ca(OH) 2 /iodoform pulpectomy compared with zinc oxide eugenol (ZOE), ZOE/iodoform, and ZOE/iodoform combined with Ca(OH) 2 at 12-month follow-up. CI: confidence interval because the ZOE/iodoform with Ca(OH) 2 did not have enough data for the comparison. There was no statistically significant difference between the two groups of studies at 6-, 12-, and ≥18-month period when using Ca(OH) 2 /iodoform compared with ZOE group (P > 0.05) for both clinical and radiographical success rates ( Figures S1-S3).
Studies were subgrouped according to the types of molars included in their studies: mandibular molars compared with maxillary and mandibular molars. We compared the effect of the type of molars in ZOE group only, because the ZOE/iodoform combined with Ca(OH) 2 had no enough data for the comparison. There were no statistically significant differences between the subgroups of studies at all follow-up periods (P > 0.05; Figures S4-S6).
The 10 studies included in the meta-analysis were either moderate-(eight studies) or high-quality (two studies; Table 2). At 6month follow-up, the two high-quality studies (Chen et al., 2017;Pramila et al., 2016) Figure S7).

FIGURE 4
Forest plot for meta-analysis of the clinical and radiographical success rates of Ca(OH) 2 /iodoform pulpectomy compared with zinc oxide eugenol (ZOE), ZOE/iodoform, and ZOE/iodoform combined with Ca(OH) 2 at ≥18-month follow-up The radiographical success rates between Ca(OH) 2 /iodoform paste and ZOE and ZOE/iodoform combined with Ca(OH) 2 at 6month period in relation to studies quality was evaluated. The ZOE showed statistically significant higher success rates in highquality studies compared with Ca(OH) 2 /iodoform (P = 0.03, OR: 0.10, and 95% CI: 0.01-0.83). However, no statistically significant difference was noticed on high-quality studies when comparing Ca(OH) 2 /iodoform to ZOE/iodoform combined with Ca(OH) 2 (P = 0.20, OR: 0.14 and 95% CI: 0.01-2.83). The high-quality studies revealed a higher statistically significant difference than the moderate-quality studies when comparing Ca(OH) 2 /iodoform to ZOE (P = 0.03) and no significant difference when comparing the Ca(OH) 2 /iodoform to ZOE/iodoform combined with Ca(OH) 2 (P = 0.20; Figure S8).  Figure 8).

| Heterogeneity
Strong evidence of heterogeneity was observed in the clinical success rates at 12 (I 2 = 45%) and ≥18 (I 2 = 58%) months and radiographical success rates at 12 (I 2 = 62%) and ≥18 (I 2 = 42%) months of followup. To explore this heterogeneity, a funnel plot was generated. At 12 and ≥18 months, both clinical and radiographical success rates on the graphs showed an asymmetry indicating that this heterogeneity may be due to chance.

| Evaluation of small study effects
Funnel plots were used for all studies together evaluating the success rates between Ca(OH) 2 /iodoform compared with ZOE, ZOE/iodoform, and ZOE/iodoform combined with Ca(OH) 2 . Absence of small study effect was found as the graphs had the shape of a funnel, and the studies were almost symmetrical around the central line at 6 months both for clinical and radiographical success rates.
Conversely, funnel plots evaluating the 12-and ≥18-month clinical and radiographical success rates on the graphs showed an assymetry, indicating the presence of publication bias (Sedgwick, 2013; Figure S9).
Egger's test was conducted to quantitatively determine asymmetry around central lines in generated funnel plots, thereby allowing us to further investigate whether small study effects were present.
No statistically significant small study effect was detected at 6 months regarding clinical and radiographical success rates (clinical P = 0.93 and radiographical P = 0.58), 12 months clinical and radiographical success rates (clinical P = 0.66 and radiographical P = 0.30), and clinical success rates at ≥18 months (P = 0.79). However, a quantitative asymmetry was observed in the funnel plot depicting radiographical success rates at ≥18-month follow-up, indicating statistically significant small study effects (P = 0.02).

FIGURE 6
Forest plot for meta-analysis of the radiographical success rates of Ca(OH) 2 /iodoform compared with zinc oxide eugenol (ZOE) and ZOE/iodoform combined with Ca(OH) 2 at 12-month follow-up within studies of high and moderate quality 3.7 | Level of evidence

FIGURE 7
Forest plot for meta-analysis of the clinical success rates of Ca(OH) 2 /iodoform compared with zinc oxide eugenol (ZOE) and ZOE/ iodoform combined with Ca(OH) 2 at ≥18-month follow-up within studies of high and moderate quality Trairatvorakul & Chunlasikaiwan, 2008). Barja-Fidalgo, Moutinho-Ribeiro, Oliveira, and de Oliveira (2011) investigated permanent teeth pulpectomy success rates and revealed that there was no difference in outcomes for maxillary or mandibular teeth. Expanding upon these results, our study determined that pulpectomy success rates in primary teeth were also not affected by tooth type.
When stratifying our meta-analysis by study quality, differing results were uncovered. We found that ZOE use was associated with a statistically significantly higher success rates than Ca(OH) 2 /iodoform in high-quality studies. This difference, however, was not present in moderate-quality studies. This difference could be a consequence of the limitations found in moderate-quality studies such as small sample size, lack of sample size calculation, the unclear design of the study, and limited time of follow-up.
According to Al-Namankany, Ashley, Moles, and Parekh (2009) and Rajasekharan, Vandenbulcke, and Martens (2015), the quality of reporting randomized clinical trials in pediatric dentistry journals was poor and inadequate for ensuring reliable and reproducible results. In addition, the CONSORT group reported that meta-analyses including low-quality randomized clinical trials may overestimate success rates of a given medical intervention by 35% in Medicine (Moher et al., 1998;Schulz, Chalmers, Hayes, & Altman, 1995). We believe that our subgroup analysis comparing success rates within studies of high and moderate quality provides additional information that remains uninfluenced by research with a high risk of bias.
This SR and meta-analysis had some limitations. For example, we observed moderate to high levels of heterogeneity across included studies. Specifically, a moderate level of heterogeneity was found in the 12-and ≥18-month follow-up. This may have stemmed from systematic differences within the studies analyzed; that is, different eligibility criteria yielding distinct patient populations, varying levels of and rationale for participant dropout, varying methods used to evaluate radiographical success rates, differences in study design (randomized vs. non-randomized clinical trials and non-blinded trials vs. single-or double-blinded trials), and variations in the clinical procedure performed (intracanal irrigation solutions, number of treatment visits, final restorative materials, type of teeth undergoing pulpectomy, and latency to follow-up).
There are no reliable methods with which to quantify the amount of clinical, radiographical, and methodological heterogeneity. Careful selection of appropriate studies is the only way to ensure the derivation of accurate inferences in meta-analyses. Despite attempts to include a large number of related studies in our analysis, our search yielded only 15 studies, two of which were deemed high-quality studies suitable for inclusion. The small number of studies included in our meta-analysis leading to substantial bias of heterogeneity (Von Hippel,

FIGURE 8
Forest plot for meta-analysis of the radiographical success rates of Ca(OH) 2 /iodoform compared with ZOE and ZOE/iodoform combined with Ca(OH) 2 at ≥18-month follow-up within studies of high and moderate quality 2015). To overcome this heterogeneity, we applied a random effects model and performed subgroup analysis; we feel that this allowed us to contrive reliable results.

| CONCLUSION
On the basis of the current study findings, we believe that due to its resorbable property, Ca(OH) 2 /iodoform is the best filling material to be used for pulpectomy in primary teeth nearing exfoliation. Conversely, either ZOE or ZOE/iodoform combined with Ca(OH) 2 is the materials of choice for pulpectomy in primary teeth need long time before exfoliation.
The clinical and radiographical success rates of Ca(OH) 2 /iodoform paste are comparable with that of ZOE in primary teeth pulpectomy up to ≥18-month follow-up.
Future clinical trials with a high-quality randomized controlled clinical trials and long-term follow-up period are needed before a reliable conclusion can be drawn as to the best pulpectomy material in primary teeth.

ACKNOWLEDGMENT
This review received no grant from any funding agency in the public, commercial, or not for profit sectors.