Prevalence of nonsyndromic oral clefts in Texas: 1995–1999

INTRODUCTION

Orofacial clefts, which includes nonsyndromic cleft lip with or without cleft palate (NSCLP) and nonsyndromic cleft palate only (NSCPO) are the fourth most common birth defect in the United States [Robert et al., 1996; Tolarova and Cervenka, 1998]. Although genetic factors are thought to play an important role in the cause of clefts different environmental factors have been implicated as well [Wyszynski, 2002]. Some studies suggest that maternal dietary intake of folic acid may help to prevent the occurrence of oral clefts, while other studies have not found this association [Menegotto and Salzano, 1991; Stoll et al., 1991; Shaw et al., 1995, 2002; Croen et al., 1998; Schaffer et al., 1998; Cooper et al., 2000; Rajabian and Sherkat, 2000; Vieira and Orioli, 2002]. Due to the established beneficial effect of folic acid in reducing neural tube defects, the FDA mandated that all grain products should be fortified with folic acid by January 1, 1998. To date, no population-base studies have evaluated the effect of folic acid fortification of grain products on the prevalence rates of nonsyndromic oral clefts.

This study was undertaken to describe the prevalence and demography of NSCLP and NSCPO in Texas between 1995 and 1999. Moreover, we examined two factors that have not been determined previously: (1) rates of offspring with NSCLP and NSCPO among Hispanic mothers who were born in the US and Hispanic mothers who were born outside of the US and (2) rates of offspring with NSCLP and NSCPO conceived before and after mandatory folic acid fortification was introduced in the US.

METHODS

Cases of oral clefts were ascertained by the Texas Birth Defects Monitoring Division (TBDMD) at the Texas Department of Health, which conducts active surveillance for birth defects in Texas. Cases of cleft lip with or without cleft palate were identified using Centers for Disease Control/British Pediatric Association (BPA) codes 749.10 to 749.29, while those for cleft palate alone were identified using BPA codes 749.00 to 749.09. All cases were reviewed by a clinical geneticist (J.T.H.). All multiple gestations (twins, triplets, etc.) and cases with chromosomal abnormalities or one or more major malformation(s) other than oral clefts were excluded from the study. Only nonsyndromic cases of oral clefts remained in the study. Cases that had minor defects such as nevus flammeus, nasal septum deviation, flattened nares, eyelid adhesion, skin tags, palmar crease, sacral dimple, mild hypertelorism, asymmetric chest, epicanthal fold, natal teeth, etc., were also considered to be nonsyndromic and were retained in the study. Cases were further subclassified as nonsyndromic NSCLP or nonsyndromic NSCPO, as these two types of oral clefts are thought to be genetically, embryonically, and etiologically different from each other.

Texas is divided into 11 Public Health Regions (PHRs) as follows: Region 1 and 2 includes Northwest Texas and the Texas panhandle; Region 3 includes Dallas, Ft. Worth, and surrounding areas; Region 4 and 5 comprise East Texas; Region 6 includes Houston, Galveston, and surrounding areas; Region 7 includes Austin and Central Texas; Region 8 includes San Antonio and surrounding areas; Region 9 and 10 comprise West Texas; and Region 11 includes Corpus Christi, Brownsville, and the Lower Rio Grande Valley (http://www.tdh.state.tx.us/bfds/images/map1.gif). Between 1995 and 1998 TBDMD ascertained cases from selected PHRs. In 1999, ascertainment of birth defects became statewide. Some of the PHRs in Texas are large and sparsely populated and thus have few births. In order to provide more precise rates, the following contiguous regions were merged and analyzed as single regions of interest: PHR 1 and 2, PHR 4 and 5, and PHR 9 and 10.

Records of infants with birth defects were linked to their birth certificates or fetal death certificates filed with the Texas Bureau of Vital Statistics, Texas Department of Health. The birth or fetal death certificate was used to obtain information for the following variables: date of delivery, sex of the infant or fetus, mother's age, race/ethnicity, parity, and county of residence at the time of delivery. In rare instances when a case did not match a birth or fetal death certificate, or when information was missing from the birth certificate, the information for that particular variable was obtained from abstraction of the medical records performed by TBDMD. Denominators for our prevalence rates were based on live births corresponding to the geographic areas in Texas that were under surveillance by TBDMD in a particular year or portion thereof.

Prevalence rates and prevalence odds ratios (POR) were calculated for NSCLP and NSCPO separately. The reported last menstrual period was used to calculate the estimated date of conception for the folic acid fortification analysis. The pre-fortification group was comprised of all births conceived on or before December 31, 1997 and all births conceived on or after January 1, 1998, which was the mandatory folic acid fortification date, were in the post-fortification group. Adjusted rates were calculated using a logistic regression model to adjust for all other variables simultaneously. All analysis was done using SPSS (version 11.5, copyright 2002, SPSS, Inc., Chicago, IL) statistical software.

RESULTS

The total number of infants born with cleft lip with or without cleft palate during the 5-year study period was 1,082. Of this total, 651 (60%) were classified as NSCLP. During the same period, 612 cases of cleft palate only were ascertained, of which 245 (40%) were classified as NSCPO. The total number of singleton births born to women residing in the PHRs covered by TBDMD between 1995 and 1999 was 1,026,868. Prevalence rates as well as the crude and adjusted prevalence odds ratios (POR) for NSCLP and NSCPO are listed in Tables I and II, respectively. PORs changed little after adjustment, therefore, only the adjusted POR are described below.

The prevalence of NSCLP was 0.67/1,000 among whites. US-born Hispanics and foreign-born Hispanics had a prevalence of 0.63/1,000 and 0.65/1,000, respectively. The prevalence of NSCLP in blacks was 0.40/1,000 live births, considerably lower than in whites (POR = 0.61, 95% confidence interval [CI] = 0.44–0.84). After adjustment, the highest rates for NSCLP were found in PHRs 9 and 10 of West Texas, (POR = 1.47, CI = 1.06–2.03, referent group PHR 3).

NSCLP was found to be less frequent in females than in males (POR = 0.61, CI = 0.52–0.71). After adjustment, neither maternal age, maternal education, nor year of birth were associated with the prevalence of NSCLP. However, a trend of increasing rates with increasing parity was observed for NSCLP, with women of parity 4 or higher having 40% higher rates of NSCLP (POR = 1.40, CI = 1.06–1.86).

Prevalence rates of NSCPO were lower than those of NSCLP with whites having a prevalence of 0.32/1,000 live births. The prevalence rate of NSCPO was 0.21/1,000 live births in US-born Hispanics and 0.14/1,000 live births in foreign-born Hispanics. Compared to whites, the prevalence in US-born Hispanics was 38% lower (POR = 0.62, CI = 0.42–0.90) and the prevalence in foreign-born Hispanics was 54% lower (POR = 0.46, CI = 0.30–0.71).

As expected, NSCPO was more common in females than in males (POR = 1.30, CI = 1.01–1.68). After adjustment, neither maternal education nor maternal age were associated with the prevalence of NSCPO. Although the rates of NSCPO were slightly higher for multiparous women in comparison to nulliparous women, no significant trend was observed.

The PORs for NSCLP and NSCPO before and after fortification of the US grain supply with folic acid are listed in Table III. These PORs were adjusted for infant's gender, and mother's age, ethnicity, education, parity and PHR. The rates were slightly lower for NSCPO (POR = 0.87, CI = 0.66–1.15) after fortification.

DISCUSSION

This is the second population-based study to present prevalence rates of oral clefts separately for the US-born and foreign-born Hispanic populations. Rates of NSCLP in US-born and foreign-born Hispanics were similar to those of whites. However, rates of NSCPO were significantly lower in both of the Hispanics groups compared with rates among whites. A similar pattern was observed in a population based study by Croen et al. [1998] of oral clefts occurring in California from 1983 to 1999. We also observed lower rates of NSCLP and NSCPO among blacks similar to what has been reported by a number of previous studies [Shaw et al., 2002].

Here, we report prevalence rates of nonsyndromic CLP and nonsyndromic CPO in Texas 1995–1999. In contrast, Croen et al. [1998] reported rates for all cases of CLP and CPO. They noted that 73.8% and 48.3% of their cases of CLP and CPO cases were isolated. Their definition of isolated cases corresponds to our definition of nonsyndromic cases. Using these percentages we estimate that the rates of isolated CLP from the data of Croen et al. [1998] were 0.78, 0.77, and 0.48 per 1,000 births for whites, Hispanics, and blacks, respectively. The estimated rates of CPO from their data are 0.35, 0.27, and 0.26 per 1,000 births for whites, Hispanics, and blacks, respectively. After this adjustment, the rates reported by Croen et al. [1998] remained higher than our rates, yet the pattern of rates across ethnic groups remained the same. Although our definition of “non-syndromic” was the same as their definition of “isolated,” we may have applied it more narrowly, classifying only 60% of CLP and 40% of CPO as nonsyndromic. This difference may explain the lower rates we report.

Protective effects of folic acid on the occurrence of oral clefts have been reported previously. Two separate studies have reported reductions of 47% and 50% in isolated cleft lip and palate and reductions of 19% and 27% in isolated cleft palate alone in population-based case control studies using folic acid containing multivitamins [Shaw et al., 1995, 2002]. Czeizel et al. [1996] also reported a beneficial effect of folic acid in a population-based case control study using daily high dose folic acid supplementation. Itikala et al. [2001] reported a 48% risk reduction in cases of cleft lip with or without cleft palate (OR = 0.52 CI = 0.34–0.80) and a 19% for cleft palate alone (OR = 0.81 CI = 0.44–1.52) among women who used folic acid containing multivitamins during the periconceptual or early postconceptual period. This was a population-based case-control study, where cases included isolated clefts as well as those clefts occurring in conjunction with other unrelated major defects. Loffredo et al. [2001] also reported similar results using a comparable study design where the beneficial effect of periconceptional vitamin use was greater for NSCLP than for NSCPO. On the other hand, Werler et al. [1999] reported a 30% risk reduction for cleft lip and palate and a 60% reduction for cleft palate only. However, a protective effect has not been reported in all populations [Shaw et al., 1991; Czeizel and Dudas, 1992; Hayes et al., 1996; Ray et al., 2003].

In our study population, there was no difference between the rates of NSCLP after fortification of grain products (POR = 0.94, CI = 0.79–1.12). A 13% decrease in the rate of NSCPO was observed in the post-fortification period (POR = 0.87, CI = 0.68–1.15), however, this should be interpreted with care as it may be due to chance. It is possible that folic acid results in a decrease in oral clefts only after serum or RBC levels of folic acid reach a threshold value. The average daily dose of folate provided by food fortification is 0.1 mg, whereas the dose in multivitamins which is 0.4 mg per day. Thus, our observation of no little or no decrease in the prevalence of oral clefts subsequent to folic acid fortification is not inconsistent with case-control studies which show a reduction in risk among women consuming multivitamins containing 0.4 mg of folic acid daily. It is also possible that the reduction in risk of oral clefts associated with the use of multivitamins in the case-control studies is due to the consumption of vitamins other than folic acid.

After adjustment for ethnicity and other demographic factors, the rates of NSCLP were elevated in West Texas (PHRs 1, 2, 9, and 10). This may reflect different environmental exposures in West Texas, a difference in ascertainment between West Texas and PHR 3 or it may be a chance finding. Analysis of rates in additional years will resolve whether NSCLP rates continue to be elevated in these regions.

There are studies that both support and refute an association between birth order and oral clefts [Menegotto and Salzano, 1991; Shaw et al., 1991; Stoll et al., 1991; Cooper et al., 2000; Rajabian and Sherkat, 2000]. Using a meta-analysis, Vieira and Orioli [2002] also found an increase in the risk of oral clefts with increasing parity. For parity of 4 or more, they reported an odds ratio of 3.12 (CI = 2.58–3.77) for NSCLP and an odds ratio of 3.35 (CI = 2.39–4.69) for NSCPO. Although our result of a 40% risk elevation for NSCLP in women of parity 4 or higher (POR = 1.40, CI = 1.06–1.86) is of a smaller magnitude, it is consistent with the direction of their results. We did not, however, observe an association between parity and an increased risk of NSCPO. Women who have a greater number of babies are generally somewhat older and have lower levels of education compared with women who have fewer babies. Thus, it is important to note that the associations we observed between parity and NSCPO are adjusted for both maternal age and maternal education.

Neither maternal age nor maternal education was an independent risk factor for NSCLP or CPO in our study. Unfortunately, we were not able to examine maternal smoking which has been observed to be a modest risk factor for oral clefts in previous studies [Wyszynski et al., 1997].

In summary, we found that when Hispanic women are compared to white women they have similar rates of NSCLP but decreased rates of NSCPO. These findings are similar to those of a previous population based study by Croen et al. [1998]. We also observed that infants conceived after the onset of fortification of grain products with folic acid in the US, had no decrease in prevalence of NSCLP and a very modest decrease in the prevalence of NSCPO which may be explained by chance. Thus, our study suggests that folic acid fortification has had little or no impact on the prevalence of nonsyndromic oral clefts in Texas.