Leprosy and pregnancy in the State of Pará : an epidemiological perspective

Introduction: A few older publications describe leprosy associated with pregnancy, a situation that has been linked to leprosy exacerbation. This study aimed to describe the detection rate of this association in the State of Pará by county and Integration Region (IR) from 2007 to 2009 via an analysis of socio demographic, epidemiological and operational indices. Methods: This was a descriptive study using information generated by the SINAN. The Detection Coefficient of the Leprosy and Pregnancy Association (DCLP) epidemiological index was constructed to help interpret the endemicity parameters. The disease was considered hyperendemic when greater than two cases per 10,000 inhabitants were identified. Results: During the study period, 149 associations were detected, with 14 hyperendemic counties: seven in 2007, five in 2008 and two in 2009. The Carajás Integrated Region displayed the highest DCLP index in the period. Eldorado dos Carajás had the single highest DCLP index (5.7/10,000 inhabitants, 2008), whereas the DCLP index in Conceição do Araguaia was very high in all three years. However, most counties displayed low or medium DCLP indices. The annual averages were 0.31 DCLP (2007), bass; 0.30 (2008), bass and 0.19 (2009), bass. The average DCLP index was 0.26, which is considered low. Three clusters of medium endemicity were identified by the average DCLP in the study period. Conclusions: The analyses indicated that the surveillance program is still unsatisfactory in Pará. The interpretation of the endemicity parameters enabled qualitative and quantitative analyses to determine the epidemiological panorama of this association. The identification of high endemicity requires further clarification.

Leprosy is a disease caused by Mycobacterium leprae and curable with multidrug therapy (MDT) 1 .Numerous studies on the transmission, clinical presentation and prognosis of leprosy corroborate the now indisputable relevance of the socioeconomic and cultural changes in the disease's epidemiological profi le [2][3][4] .The association of unfavorable socioeconomic and cultural factors with disease sequelae demonstrates the vulnerability of the affected population, who lose quality of life because the current public health treatment options do not yet permit the reversibility of the leprosy stigma 5 .
In Brazil, although the detection rates are declining steadily, year by year, the northern region still remains as the most endemic, both in terms of general leprosy indices and cases associated with pregnancy 6 .
Leprosy control relies on detailed and accurate information to support the plan and design of intervention strategies.The implementation of the Notifi able Diseases Information System (SINAN) occurred heterogeneously in the Brazilian states, and a software-based system for the SINAN was not implemented until 1998 7 .
The National Plan for the Elimination of Leprosy (PNEH) is a priority of the Health Pact of 2006 and currently relies on the new cases detection index, which replaced the punctual prevalence index 8 .Due to the vast Brazilian territory, leprosy is heterogeneously distributed in the country.Whereas the leprosy prevalence is below one case per 10,000 inhabitants in the south, in several states of the northern, northeastern and center-western regions, leprosy reaches hyperendemic detection coeffi cients (DC).The State of Tocantins has the highest DC in Brazil 9,10 .
T he spatial-temporal distribution of leprosy can be visualized with coropletic and thematic maps created with geoprocessing techniques, which permits the synthesis of epidemiological relationships occurring in time and space, generating useful data on leprosy and other diseases 11 .
Although the association between leprosy and pregnancy appears uncommon, the attendant complications of this relationship are not 12 .In the final gestational trimester, a depression of cellular immunity occurs, which can result

METHODS
in reactional states such as Lucio's phenomenon (a type II reaction); the additive effect of this reactional state to the characteristic vascular changes of pregnancy results in a higher probability of severe necrotizing vasculitis, typically in the lower limbs.In the postpartum period, a suppression of humoral immunity occurs, with intensifi cation of rapid type I reactions, resulting in irreversible nerve necrosis 13 .The treatment of type II reactions is based on thalidomide.Use of this drug requires particular vigilance in women of childbearing age, due to its known teratogenicity 14 .
Several challenges exist in adapting national and state statistics to the goal of eliminating leprosy within the time limits established by the National Plan for the Elimination of Leprosy (NPLE) 8 .These challenges include late diagnosis, patient noncompliance with the treatment and failures in the registration and notifi cation fl ows.A clearer view of these problems is necessary to enable review strategies and adoption measures to eliminate leprosy as a public health problem.The gravity of leprosy and its disruptive sequelae render essential an exploration and analysis of the factors behind pregnancy and leprosy in the State of Pará 15 .More than sixty years ago, Ryrie wrote that among the comorbidities during pregnancy, leprosy might be one of the few systemic diseases for which the interaction is unilateral.Leprosy does not appear to have an effect on pregnancy development, but the characteristic immune, hormonal and metabolic alterations of pregnancy do affect the course of leprosy 16 .
Data from Duncan et al. 17 indicate a clear association between the postpartum period and the developing of neuritis.They reported patients with evident neuritis, namely decreased sensory and/or motor nerve function without pain or sensitivity 17 .
The repercussions of leprosy surpass the limits of health institutions, as their resultant deformities and disability entail withdrawal from the workforce and the need for specialized technical resources carrying economic burdens, in addition to the millennia-old cultural stigma of sin and divine punishment related to the condition, which leads to social segregation and a loss of quality of life 18 .The need for data on the disease mandates a review of current epidemiological indices, which are not suffi ciently clear on situations of collective risks that may explain how this disease manifests itself socially 4 .
This study sought to analyze the epidemiological situation of the association between leprosy and pregnancy, describing socio-demographic variables, operational and epidemiological indices and the construction of the detection coeffi cient of the leprosy and pregnancy association (DCLP) to obtain the necessary interpretation parameters for this endemicity association 19 .
We chose to conduct a retrospective longitudinal and descriptive study, using documentary research through the SINAN reporting forms from 2007 to 2009, based on municipalities (counties) and the twelve Integration Regions (IR) of the State of Pará.Data were also collected from the State of Pará Department of Public Health (SESPA) and the Brazilian Institute of Geography and Statistics (IBGE), the Ministry of Health Surveillance System (SVS/MS) and the DATASUS (Unifi ed Health System).Electronic Database.The study population consisted of pregnant women 12 to 49 years of age reported with a leprosy diagnosis in the three-year study period.The described variables were age, clinical presentation, education level, operational classifi cation, number of skin lesions, input mode, number of affected nerves, type of discharge, method of case detection, bacilloscopy results, initial therapeutic regimen, number of contacts registered and reactional episodes.The described operational indices were percentage of contacts investigated, percentage of patients evaluated for disability at diagnosis, number of patients cured in the period, percentage of patients assessed for disability at healing and percentage of cases with grade II disability.An epidemiological index, the DCLP, was constructed, as well as a method for interpreting the endemicity parameters, using percentile statistical analysis.
For the contingency table statistical analysis, the G test with Williams' correction was applied, with a two-tailed p <0.05 considered statistically signifi cant.To calculate the DCLP, the numerator is the pregnant leper total, and the denominator is the total number of women aged 12 to 49 years, multiplied by 10,000 inhabitants 19 .

DCLP =
Number of pregant lepers x 10,000 Total number of women aged 12 to 49 years To estimate the endemicity patterns of the leprosy-pregnancy association, all DCLP indices calculated were, from start, arranged in ascending order for the studied counties.The State of Pará has 143 counties.Considering that the study was conducted for a period of three years, we can state that 429 coeffi cients of detection were observed along this time.The number 84 indicates that only the positive coeffi cients were considered, and the zeros were not counted.Thus, out of 429 coeffi cients, only 84 are positive (i.e., greater than zero).After calculating the coeffi cients, the percentile method was used because there are fi ve endemicity patterns: hyperendemic, very high, high, medium and low. Figure 1 shows that four percentiles were used (P20, P40, P60 and P80).The DCLP values corresponding to these percentiles were then calculated.These four percentiles divide the ordered series of coeffi cients into fi ve groups with an equal number of observations as previously proposed by Palacios et al. 20 .Each group corresponds to a range that characterizes the disease endemicity parameter (low, medium, high, very high and hyperendemic).
Once calculated, the DCLP values were interpreted according to the predefi ned parameters.Thematic and coropletic digital maps were then generated, using ArcView 3.2 software, integrated with Corel Draw, version 1.5, so as to depict the spatial distribution of the DCLP values visually.In this study, all p values were greater than 0.05, indicating no statistical evidence against the hypothesis that leprosy cases in pregnant women were distributed evenly among the categories of the following selected variables.

RESULTS
With regard to the age and race of the 149 women tallied, 104 (69%) were aged 20-39 years and 84 (56%) were members of the pardo racial classifi cation.As for schooling, no signifi cant predominance was identified, although most subjects had completed elementary and high school education; four were illiterate.Regarding the clinical form of the disease, 49 (32%) cases were indeterminable and 45 (30%) dimorphic.Of the 149 cases studied, 28 (18%) were unclassifi able.Regarding the operational classifi cation, the paucibacillary form was observed in 84 (56%) cases, and the multibacillary form was observed in 65 (44%) cases.Single skin lesions appeared in 74 (49%) cases, whereas 49 (32%) cases had two to fi ve lesions.The input mode revealed 124 (83%) new cases.Regarding the number of affected nerves, the not informed column was predominant, comprising 102 (68%) of the 149 reported cases.There were 92 (61.7%) cases discharged following cure, and 21 (14.1%)patients abandoned treatment.Regarding the detection of cases, 73 (48%) pregnant women with leprosy sought health services spontaneously, followed by 34 (22%) referral cases, whereas 25 (16%) cases were registered as ignored.The bacilloscopy results were ignored in 98 (65%) cases and unrealized in 37 (24%) cases, meaning that 135 (90%) of the total of 149 cases were noticeable.Regarding the initial therapeutic regimen, 89 (59%) women began paucibacillary multidrug therapy (MDT/PB) and 59 (39%) were given multibacillary multidrug therapy (MDT/ MB).In addition, two to fi ve registered contacts predominated in 92 (61%) cases, whereas 30 (20%) cases registered more than fi ve contacts.Of the total, 103 (69%) cases were reported as being without reaction, and 36 (24%) cases were determined to be not reported.In addition, seven (4%) cases with type I reactions and two cases with type II reactions were reported (Table 1).
The operational index percentages were recorded for the contacts investigated, the number of patients disabled at diagnosis, the number of patients cured during the study period, the number of patients disabled at disease resolution and the number of cases with grade II disability.
The DCLP annual averages were 0.31 (2007), bass; 0.30 (2008), bass and 0.19 (2009), bass.The average DCLP index was 0.26, which is considered low.Reductions in the endemic DCLP indices were noted from over the three-year period.Namely, from 2007 to 2008, several counties were hyperendemic: two counties in the Tocantins IR, two in the Carajás IR, one in the Capim IR and two counties in the Xingu IR.In contrast, from 2008 to 2009, no county demonstrated a hyperendemic pattern, but six counties had a very high endemicity pattern: one in the Xingu IR, two in the Capim IR, two in the Araguaia IR and one in the Lago de Tucuruí IR.No county remained hyperendemic for two consecutive years, but Conceição do Araguaia (Araguaia IR) displayed a very high standard endemicity for all three years of the study.Eldorado dos Carajás (Carajás IR) and Anapu (Xingu IR) were the counties with the highest overall average DCLP indices over the three-years study period, and Eldorado dos Carajás had the highest single DC index of the study, with 5.97 cases per 10,000 inhabitants in 2008 (Table 2).

DISCUSSION
According to Veiga 21 , health analysis incorporates technologies from other fi elds of knowledge that apply to environmental and socioeconomic issues.Thus, a higher density of cases of the pregnancy-leprosy association was noticed in the Xingu, Araguaia, Carajás, Lago de Tucuruí, Capim and Marajó IR.Except for Marajó, the counties mentioned are notable for having concurrent economic development projects such as mining and hydroelectric power dams, in addition to intensifi ed industry and agriculture sectors 22 .The aforementioned projects have been attracting large migrant populations, which contribute to the epidemiological context of leprosy, given the prevalence of unplanned urbanization that results in poverty belts and their sequelae.Among the maladies associated with unplanned urban settlements are malnutrition, lack of disease immunity and insuffi cient living conditions, such as a lack of basic sanitation, housing and food, all of which favor the hidden prevalence and transmission of disease.
Focusing on the mean values over the three-year study, we identifi ed some clusters of medium endemicity encompassing the Tapajós, Xingu, Araguaia, Carajás, Lago de Tucuruí, Baixo Amazonas, Marajó, Capim and Caetés Integration Regions.Tapajós, for example, is a poor region with the lowest Gross Domestic Product (GDP) of the state, in addition to poor access to education and health.The Xingu IR has been experiencing economic growth and the installation of large projects such as the Belo Monte Dam, which has been attracting people from other areas, generating unplanned urbanization and settlements in the peripheries, along with public insecurity and increasing health problems, particularly contagious diseases.The Araguaia IR has solid economic sectors -such as farming, agrobusinesses and temporary crops -that permanently attract new settlers.The Lago de Tucuruí IR also has projects such as the Tucuruí  hydroelectric power dam.The Capim IR has been experiencing rapid but disorderly economic growth starting with the construction of the Belém-Brasília highway and federal project funding, resulting in large areas of deforestation and, fi nally, human settlements 22 .These characteristics together generate an epidemiological scenario favoring contagious diseases, such as leprosy, in which humans 4 are both the transmitter and the reservoir.Barcarena (Tocantins IR) was the county with the highest number of cases in the study period (13 of 149 cases).This IR includes the State of Pará, with the third largest GDP, with Barcarena contributing 67.2% of the GDP in the said region.Approximately 10% of the state population lives in this area of Pará, due to job opportunities brought about by the industry sectors of aluminum production and processing and the production of conducting cables (main), according to data from the State of Pará Department of Planning, Budgeting and Financing (SEPOF) 22 .These factors favor human settlements and the formation of impoverished areas (Figure 2).
The age group found to have the highest incidence of the pregnancy-leprosy association was the 20-to 39-yearold age group, which includes workers with the most intense professional activity levels.The interpersonal contact promoted by professional activities is one of the main means of transmitting leprosy 4 , particularly when associated with pregnancy.In this study, 2.7% of the patients were younger than 15 years of age, which in the WHO view indicates that the disease detection rate in this age group still displays secular trends and recent transmission 4,23 .
As for the patients' racial background, Pardo (various shades of racially mixed individuals with some African ancestry) predominated among the patients in this study, but this fi nding does not yet allow the conclusion that race infl uences disease occurrence, as 69.5% of the population of Pará is Pardo 6 .These results concur with the fi ndings of previous studies 24 .
Regarding the clinical form, the most prevalent form was indeterminable (32.89%), followed by dimorphic (30.2%).What is surprising about the dimorphic form is its multifaceted presentation that requires that the diagnosis be made by a specialist, but as any individual can notify health authority about cases, numerous misconceptions can result in under-or overnotifi cation.The dimorphic MB form is considered of higher transmission potential.These results thus confi rm the view of pregnancy as a worsening or triggering factor for leprosy 13,25 .These fi ndings contradict the literature when compared with general leprosy, where 59% of cases are multibacillary, at the expense of virchowians 4 .The presence of cases with ignored or non-informed operational classification (18.8%) allow the inference that the professionals responsible for diagnosis and reporting lack certainty in classifying the clinical form 4 .Regarding the number of skin lesions, patients with a single lesion (74 [56.4%] cases) predominated, followed by patients with two to fi ve lesions (49 cases), a distribution that confl icts with the incidence of MB forms identifi ed clinically, indicating that the operational classifi cation can lead to the misdiagnosis of clinical multibacillary disease as paucibacillary.
Most cases had no nerve damage, or the notification forms contained no information on this aspect (68.5%).This percentage precludes an accurate assessment of nerve damage and reiterates the need for more qualifi ed personnel to fi ll out the reporting forms 6,23 .Moreover, it is clear from the data that a dermatoneurologic examination is not part of prenatal care, even in endemic regions such as the Brazilian north, northeast and center-western areas.
The majority of cases reported during the study period were new cases (83.2%), which, compared to discharged following cure, accounting for 61.7% of cases, points to a worrying situation of treatment failure.The number of abandoned treatments deserves attention (14.1%).According to the Brazilian Ministry of Health 26 , the cure rate is one of the disease control indices, with the parameters used: good (>90%), regular (75 to 89%) and precarious (<75%).Discharges following cure involving leprosy associated with pregnancy in the State of Pará are classified under the precarious parameter.The Caetés and Xingu IRs were classifi ed as good in this regard, whereas the Baixo Amazonas and Tapajós IRs were found to fall into the regular category.The other IRs were classifi ed as precarious.
The disease was mainly detected by spontaneous demand (48.99%) for treatment.It is known that women are more concerned about their appearance, and the appearance of lesions bothers them more than it does men, a factor that favors their seeking out of health facilities 4 .
Concerning the results of lymphatic bacilloscopy, the fi nding that 65.8% of cases fi t the classifi cation ignored -added to the 24.8% of cases classifi ed as unrealized -renders the related information for more than 90% of the cases inconclusive.This exam is important in monitoring the disorder because it can aid in defi ning the operational form 27 , and not performing the exam certainly harms the defi nition of a reliable profi le of the association under study.
Most individuals reported two to fi ve contacts (78.5%), similar to the numbers of general leprosy (73.1%) 6 .According to Santos 24 , the risk of developing the disease is 5 to 10 times higher if a family member has already manifested the disease, so periodic examinations of contacts are necessary because only a small number of cases are detected when examined at the time of the diagnosis of the index case (Table 1).
In pregnancy, due to the higher cellular immunity depression in the fi nal trimester, if leprosy is present, the risk of leprosy reactions appearing is higher, especially the type II reactions (i.e., Lucio's phenomenon).In the postpartum period, humoral immunity is depressed, and the swift onset of type I reactions is predominant, resulting in irreversible nerve damage 13,25 .
In this study, there were no reports of any such reactions, which is contrary to previously published data 13,25 ; the absence of such information was observed in 24.2% of cases.
Regarding the assessment of patients for the presence of physical disability at the time of diagnosis and cure in this study, according to the Ministry of Health parameters 26 , the State of Pará displayed worsened performance from time of diagnosis to the time of cure, falling from the regular to the precarious parameter, except for the Guamá and Baixo Amazonas IRs, which increased from the regular to the good parameter.These results denote delayed diagnosis or the negative infl uence of pregnancy on the evolution of the disease.In reference to the occurrence of grade II disability at diagnosis, Pará was ranked as medium, and the Araguaia, Capim and Tapajós IRs were classifi ed as high.The presence of grade II disability at diagnosis is an indicator of the late detection of the association, and therefore, the high rates found in some IRs suggest the failure of surveillance measures, pointing to delayed diagnosis and contributing to an increasing number of cases of concomitant leprosy and pregnancy in the State of Pará.
In conclusion, the construction of an index for the qualitative and quantitative analysis of the leprosy-pregnancy association was useful because it made it possible the generation of an epidemiological overview of said association in the State of Pará.The highest density of the association occurrence, as well as of general leprosy, occurred in areas with active economic growth and/or implementation of major projects, thus confi rming the onerous socioeconomic burden of this disease, with or without pregnancy.The same can be inferred about the most prevalent age group, found to be the more economically and professionally active.The discrepancy seen between the operational classifi cation and the clinical form suggests that the operational classifi cation can lead to a misdiagnosis of clinical multibacillary disease as paucibacillary disease, which can then also lead to mishandling of the associated cases.
Better professional training is imperative to minimize surveillance and treatment diffi culties, some found in this study, which added to the unsatisfactory results of the epidemiological and operational indices, thus suggesting that the surveillance measures require adjustments.Because we are discussing an association between a disease and a physiological state such as pregnancy -and taking into consideration that northern Brazil is the most endemic region of the Americas, followed by the Brazilian northeast and center-west -a mandatory clinical dermatological examination in prenatal care would be useful, especially in those regions, as the clinical and epidemiological examination is the gold standard for the diagnosis of leprosy.Last but not least, it is essential to conduct more studies on this association because the immunological and clinical-epidemiological aspects of this peculiar interaction still require a great deal more explanation.This reality, when added to the lack of publications on the subject, indicates that this study can also serve as a reference.This research group will briefl y present the DCLP indices of all Brazilian regions, by municipality, from 2009 to 2011.

Palácios
VRCM et al -Leprosy and pregnancy detection rate

FIGURE 1 -
FIGURE 1 -Resume of the calculation of the parameters of endemicity of leprosy associated with pregnancy.Source: Palácios, 2012.DCLP: detection coeffi cient of the leprosy and pregnancy association.

Figure 2
demonstrates the spatial distribution of the concordant cases of leprosy and pregnancy in the State of Pará during the study period.

TABLE 1 -
Results framework for the distribution of cases according to the categories of the selected variables, from 2007 to 2009 in the State of Pará, Brazil.

TABLE 2 -
The highest Detection Coeffi cient of the Leprosy and Pregnancy Association by county, State of Pará, from 2007 to 2009.