The study objective was to evaluate, through a meta-analysis, the impact of primary palatoplasty on the sagittal maxillary and mandibular relationship among patients with complete unilateral cleft lip and palate (UCLP). Electronic database and hand searches were performed. Controlled clinical trials involving non-syndromic UCLP patients were included. Selected papers had to include a group of patients undergoing lip and palate repair and a group undergoing lip repair only. Data heterogeneity was demonstrated and individual means, standard deviations, and sample sizes were collected and summarized using a random effects model meta-analysis. Although six articles were selected for the systematic review, only four were included in the meta-analysis due to large discrepancies in the standard surgical protocol. Only one variable assessing the intermaxillary relationship (A point–nasion–B point; ANB), maxillary position (sella–nasion–A point; SNA), and mandibular position (sella–nasion–B point; SNB) was common among the selected studies. No significant differences in SNA and SNB were indentified between patients undergoing lip surgery alone and those undergoing lip and palate surgery. Evaluation of ANB showed a small statistical standard mean difference of 0.36°. Impaired maxillary sagittal growth, observed in patients with UCLP, appears to be a basic consequence of lip surgical repair. Additional changes to the maxilla and mandible produced by palatal repair are minor. Methodologically rigorous controlled studies are needed to provide a stronger evidence-based basis for the surgical management of patients with UCLP.
While primary repair of cleft lip and palate has not been shown to have a significant influence on mandibular morphology, maxillary growth deficiencies are very common in patients with unilateral cleft lip and palate (UCLP) who have undergone an early surgical intervention.
The most frequently adopted surgical protocol in the treatment of cleft considers primary palatoplasty to be the major reason for maxillary growth impairment in subjects with UCLP. This is corroborated by several studies that have compared individuals with operated UCLP vs. unoperated patients. However, some studies have shown that the cumulative maxillary growth disturbance attributable to lip and palate repair is not significantly worse than that determined by lip repair alone. This suggests that lip repair may be the most important factor in maxillary growth disturbance in patients with UCLP.
This comprehensive review was undertaken to review the available evidence regarding which of these surgical procedures has a greater effect on maxillary and mandibular growth. The answer to this clinically relevant question should help us to develop better treatment strategies for patients with cleft lip and palate thereby improving both functional and well-being outcomes.
The PRISMA checklist was utilized as a reporting guide.
This meta-analysis was registered at PROSPERO (the international prospective register of systematic reviews) under registration code CRD 42012003360.
For eligibility, all articles had to have compared patients with complete UCLP who had undergone lip repair followed by palate repair to patients who had undergone lip surgery and no palate repair.
The databases used were PubMed, Cochrane Library, ScienceDirect, SciELO, and BIREME. A partial grey literature search was undertaken using Google Scholar. A hand search of the references of selected articles was also carried out to identify any article that could have been missed in the electronic database searches.
Key words ‘cleft lip palate’ were searched in combination with ‘growth’, ‘surgery’, and ‘repair’. All references were managed using reference manager software (RefWorks) and duplicate hits were removed. Search limits were the following: controlled clinical trial, prospective or retrospective, systematic review, meta-analysis, with human samples, no language restriction, from 1960 to the end of the search on 4 September 2013.
The search was performed independently by two researchers (LMB and RCA). If the title and/or abstract appeared to fulfil the inclusion criteria, the article was selected for full retrieval. All abstracts had to mention patients with UCLP and a group of patients undergoing lip surgery only and a group of patients undergoing both lip and palate surgery. Additionally, all studies had to have used a lateral cephalometric analysis. Once full articles were obtained, a second selection stage was executed in which the same set of criteria was applied. Confirmation of inclusion was sought in cases where the title/abstract was misleading due to the limited description contained within them.
Data collection process
Both researchers retrieved the required information separately. The information selected was then checked jointly and any disagreements resolved. If necessary a third author (DN) was involved in the final decision.
The information retrieved from the final selection of articles included the following: author(s), year of publication, recruitment process (random, consecutive, convenience), sample size, inclusion criteria, country of origin, age at the time of surgery, error of method, sample matching (age, gender, origin, type of cleft), surgical technique, confounding factors, analysis, and the pertinent data.
Risk of bias in individual studies
A methodology checklist was applied to analyze and quantify the risk of bias in the studies included.
Individual means and standard deviations (SD) were collected from the lateral cephalometric measurements.
Synthesis of results
The cephalometric measurements were pooled through several meta-analyses applying a random effects model. This modelling was used because of the expected heterogeneity of the samples included. Comprehensive Meta-Analysis software was used to perform the statistical analysis. All forest plots and funnel plots, as part of the meta-analysis, were also produced by the software.
Risk of bias across studies
A funnel plot was generated using the same software to verify publication bias.
Initially, 4995 articles were screened. Only 14 of them appeared to have fulfilled the inclusion criteria based on the information provided in the abstracts. Full copies of the articles were retrieved. After reading the full texts, eight articles were excluded because they did not match the primary inclusion criteria, i.e. a direct comparison of patients with UCLP undergoing surgery to the lip and palate with patients undergoing only lip repairs. Six papers were included in the qualitative synthesis, however only four were included in the meta-analysis ( Fig. 1 ). The studies by Mars and Houston and Liao and Mars were excluded from the statistical analysis due to methodological issues. The first study did not record the mean age at lip repair of the sample; in the second study, the mean age at lip surgery was 7 years for patients who had undergone lip repair only and 1 year for patients who had undergone lip and palate repair. The studies included in the meta-analysis reported the following mean age at lip repair for the patients who had undergone a lip operation only (OL): 9.5 months, 44 months, 9 months, and before 24 months. The patients who had undergone lip and palate surgery (OLP) had lip surgery at a mean age of 5.5 months, 27 months, 9 months, and before 24 months. For this group, palate surgery was performed at 20 months, 54 months, 38 months, and before 36 months.
Sample sizes ranged from 10 subjects to 47 subjects. All patients included in the samples had UCLP. A group of patients who had undergone operations on both the lip and palate (OLP) was compared to another in which patients had undergone operations on the lip only (OL).
The authors used different surgery techniques. For lip repair, the Millard technique, Tennison technique, or rotation-advancement was used. However, patients had sometimes presented to the treatment facility with a lip repair and no clear indication of the surgery technique that had been applied. In other cases, multiple techniques were used. Palate repair was done using the Oxford method, a vomerine mucoperiosteal flap and palatal mucoperiosteal flap, by mucoperiosteal pushback, or by two-flap technique. One article only informed that multiple techniques were used, and another provided no description of the technique used. Five studies stated that the patients in the samples had not undergone bone grafting surgery and one had no information on whether bone grafting surgery had been done or not.
All articles evaluated lateral cephalometric measurements to assess maxillary and mandibular size and/or relative positioning. The lateral cephalograms analyzed were obtained at the age of 13.78 years, 19.75 years, 20.1 years, 23 years, and 26.5 years for the OL group, and 14.64 years, 16.4 years, 18.58 years, 19 years, 20.8 years, and 20.41 years for the OLP group. The mean age at lip repair for the patients who had undergone surgery to the lip only (OL) in the selected studies was 9 months, 9.5 months, 44 months, and before 24 months. The patients who had undergone lip and palate surgery (OLP) had lip surgery at a mean age of 5.5 months, 9 months, 27 months, and before 24 months. For this group, palate surgery was performed at 20 months, 38 months, 54 months, and before 36 months, respectively.
Risk of bias within studies
The methodology checklist ( Table 1 ) showed an overall moderate risk of bias; some studies presented a higher risk of bias than others.
|Mars and Houston, 1990||Kupucu et al., 1995||Capelozza Filho et al., 1996||Liao and Mars, 2005||Li et al., 2006||Chen et al., 2012|
|1. The study sample represents the population of interest with regard to key characteristics, sufficient to limit potential bias to the results|
|a. Are the source population or the population of interest adequately described with respect to key characteristics?||Yes||Yes||Yes||Yes||Yes||Yes|
|b. Are the sampling frame and recruitment adequately described, possibly including methods to identify the sample, period of recruitment, and place of recruitment?||No||Yes||Yes||Yes||Yes||Yes|
|c. Are inclusion and exclusion criteria adequately described?||Yes||Yes||Yes||Yes||Yes||Yes|
|d. Is participation in the study by eligible individuals adequate?||Yes||Yes||Yes||Yes||Yes||Yes|
|e. Is the baseline study sample adequately described with respect to key characteristics?||No||Yes||Yes||Yes||Yes||Yes|
|2. The prognostic factor of interest is adequately measured in study participants, sufficient to limit potential bias|
|a. Is a clear definition or description of the prognostic factor(s) measured provided?||Yes||Yes||Yes||Yes||Yes||Yes|
|b. Are continuous variables reported, or appropriate cut-off points (that is, not data-dependent) used?||No||Yes||Yes||No||Yes||Yes|
|c. Are the prognostic factor measured and the method of measurement valid and reliable enough to limit misclassification bias?||Yes||Yes||Yes||Yes||Yes||Yes|
|d. Are complete data for prognostic factors available for an adequate proportion of the study sample?||No||Yes||Yes||Yes||Yes||Yes|
|e. Are the method and setting of measurement the same for all study participants?||Yes||Yes||Yes||Yes||Yes||Yes|
|f. Are appropriate methods employed if imputation is used for missing data on prognostic factors?||No||Unclear||Unclear||No||Unclear||Unclear|
|3. The outcome of interest is adequately measured in study participants, sufficient to limit potential bias|
|a. Is a clear definition of the outcome of interest provided, including duration of follow-up?||Yes||Yes||Yes||Yes||Yes||Yes|
|b. Are the outcome that was measured and the method of measurement valid and reliable enough to limit misclassification bias?||Yes||Yes||Yes||Yes||Yes||Yes|
|c. Are the method and setting of measurement the same for all study participants?||Yes||Yes||Yes||Yes||Yes||Yes|
|4. Important potential confounders are appropriately accounted for, limiting potential bias with respect to the prognostic factor of interest|
|a. Are all important confounders, including treatments, measured? Are clear definitions of the important confounders measured provided?||No||Yes||Yes||No||Yes||Yes|
|b. Is measurement of all important confounders valid and reliable?||No||Yes||Yes||No||Yes||Yes|
|c. Are the method and setting of measurement of confounders the same for all study participants?||Unclear||Yes||Yes||Unclear||Yes||Yes|
|d. Are appropriate methods employed if imputation is used for missing data on confounders?||Unclear||Unclear||Unclear||Unclear||Unclear||Unclear|
|e. Are important potential confounders accounted for in the study design?||No||Unclear||Unclear||No||Unclear||Unclear|
|f. Are important potential confounders accounted for in the analysis (that is, appropriate adjustment)?||No||Unclear||Unclear||No||Unclear||Unclear|
|5. The statistical analysis is appropriate for the design of the study, limiting potential for the presentation of invalid results|
|a. Is the presentation of data sufficient to assess the adequacy of the analysis?||Yes||Yes||Yes||No||Yes||Yes|
|b. Where several prognostic factors are investigated, is the strategy for model building appropriate and based on a conceptual framework or model?||Yes||Yes||Yes||Yes||Yes||Yes|
|c. Is the selected model adequate for the design of the study?||Yes||Yes||Yes||Yes||Yes||Yes|
|d. Is there any selective reporting of results?||Yes||Yes||Yes||Yes||Yes||Yes|
|e. Are only pre-specified hypotheses investigated in the analyses?||Yes||Yes||Yes||Yes||Yes||Yes|
Results of individual studies
Regarding the maxillary position (sella–nasion–A point (SNA) angle), Mars and Houston, Capelozza Filho et al., Kupucu et al., Li et al., and Chen et al. showed no statistical difference between patients with UCLP who had undergone a lip repair only and those who had undergone both lip and palate surgery. Only Liao and Mars showed significant differences in maxillary position between these tested groups ( P = 0.002). Mandibular position (sella–nasion–B point (SNB) angle) was not different between the OL and OLP groups in any study. In three of the six studies evaluated in the systematic review, the intermaxillary relationship (A point–nasion–B point; ANB) was found to be significantly different between the OL and OLP groups.
Risk of bias across studies
Despite the small number of articles included in the review, publication bias was tested. Article distribution in the funnel plot did not show asymmetry, and therefore no publication bias was found for SNA, SNB, or ANB angles ( Fig. 2 ).
Synthesis of results
Ultimately only four studies could be included in the meta-analysis. The necessary data for SNA, SNB, and ANB angles were extracted ( Table 2 ) and pooled ( Table 3 ). On meta-analysis, no significant differences were found for SNA and SNB angles between patients with UCLP who had lip and palate surgery (OLP) and those who had lip surgery only (OL), as demonstrated by the forest plots ( Fig. 3 ). Pooled data from all articles for the ANB angle showed statistically significant differences between the OL and OLP groups ( Fig. 3 ).
|Kupucu et al., 1995||10||74.7||3||74.7||3.1||0||2.2||30||76.3||3.5||76.9||4.7||−0.5||3.8|
|Capelozza Filho et al., 1996||23||77.2||3.8||77.2||3.9||0||3.2||35||76||5.6||76.8||3.8||−0.8||4.6|
|Li et al., 2006||35||73.64||4.5||77.34||4.9||−3.7||4.0||47||73.3||5.5||75.1||4.6||−1.81||3.1|
|Chen et al., 2012||15||79.25||4.9||78.98||3.2||0.2||4.3||16||75.4||6.5||79.6||6.4||−4.17||5.1|