Tuberculosis prevalence in renal transplant recipients : systematic review and meta-analysis

Os agravos crônicos, nas últimas décadas, ganharam destaque tanto no meio científico quanto no contexto da sociedade geral. Com o desenvolvimento tecnológico, já foram desvendados a fisiopatologia e os fatores de risco de muitas dessas doenças, foram descobertos métodos diagnósticos, formas de tratamento e de prevenção, porém, a diminuição dos Prevalência de tuberculose em transplantados renais: revisão sistemática e meta-análise


IntroductIon
In the last decades, chronic diseases have gained the attention of the scientific community and society in general.Technological development has helped elucidate the pathophysiology and the risk factors connected to many chronic ailments, in addition to allowing the introduction of diagnostic methods, treatment modes, and prevention strategies.However, reductions on the number of new cases of disease did not occur at the same pace of such developments. 1,2his fact, when considered in conjunction with increases in life expectancy, brings about a new context in which two or more diseases coexist in one individual. 3uberculosis (TB) is an infectious disease described in humans since classical antiquity, whose detection and treatment are subjects of the interest of health care workers.][6] Additionally, in the specific case of subjects with AIDS, problems with alcohol abuse, malignant diseases, and on immunosuppressants -as organ transplant patients, the risk of developing TB is increased. 5Thus, the prevalence of TB among individuals submitted to organ transplants is higher than in the general population. 7The prevalence of TB is determined mainly by the epidemiological risk of each country, and has been estimated to be 37 times higher among kidney transplant patients than in the general population living in endemic areas.7-9This systematic review and meta-analysis aimed to assess the prevalence of tuberculosis in renal transplant patients.

Method
The following search strategy was adopted to find studies on the prevalence of TB in individuals submitted to kidney transplant: databases DOI: 10.5935/0101-2800.20130033 MedLine and LILACS were searched for papers published between January of 2000 and December of 2011 with the following string: tuberculosis AND kidney transplantation (term "Medical Subject Headings" (MESH) and free text) for MedLine; and 'tuberculose E transplante renal' (term 'Descritores em Ciências da Saúde' (DeCS) and words) on LILACS.No restrictions were applied to the languages in which the studies were written.
The references of the selected papers were also reviewed.
Studies including only individuals under the age of 18 years, patients managed with renal replacement therapies (RRT) other than kidney transplantation, and subjects who had other organ transplants were excluded.
The search and selection of papers was carried out by two reviewers and differences of opinion were resolved by a third reviewer.Figure 1 describes the number of included papers in each step of the selection process: • 1 st step: papers listed in search results were reviewed for duplicates and organized by a reviewer (R.B.); • 2 nd step: two reviewers (R.B. and G.T.) independently analyzed the titles of the papers and excluded the studies meeting the exclusion criteria described above (non-original papers or studies enrolling only subjects under the age of 18 years, individuals on RRT other than kidney transplant, and subjects who had other organ transplants); • 3 rd step: two reviewers (R.B. and G.T.) independently analyzed the abstracts of the papers selected after the second step and excluded the ones not mentioning descriptions of characteristics/ adverse events experienced by the studied population after kidney transplantation.Disagreements between the two reviewers were resolved by a third reviewer (M.E.); • 4 th step: the papers selected on the third step were retrieved and reviewed independently (R.B. and G.T.).Papers reporting prevalence of TB in individuals submitted to kidney transplants and studies reporting data that allowed the calculation of prevalence rates were selected.Disagreements between the two reviewers were resolved by a third reviewer (M.E.).The references of the papers selected on the fourth step were reviewed, and studies featuring keywords tuberculosis, kidney transplantation, tuberculosis, renal transplant were screened based on the steps described above.
A protocol was defined to extract data from the full texts of the papers.Data extraction was performed by two reviewers (R.B. and G.T.) and disagreements were resolved by a third reviewer (M.E.).The following data were extracted: year of publication; period of data collection; site where study was carried out; study design; data sources; number of centers included; sample size; number of cases of TB and/or prevalence of TB.The prevalence of tuberculosis in the population in general of each country represented in the review was obtained from the registers of the World Health Organization (WHO). 4The authors of the selected studies did not have to be contacted for additional information.
Studies using the following criteria to diagnose TB were included: demonstration of alcohol-acid resistant bacilli (AARB) in individuals with suspected disease; growth of bacilli in culture samples; histopathology tests showing AARB or granulomatous inflammation; satisfactory response to treatment in patients with typical findings in imaging or fever of unknown origin with negative results in other tests.No distinctions were made between the clinical manifestations of TB in the studies.Individuals with pulmonary, extra pulmonary, and concomitant pulmonary and extra pulmonary tuberculosis were included.
Case-control studies were not excluded from this review, as they presented reference population data which enabled the calculation of prevalence rates.A study10 described TB prevalences for two separate populations (individuals submitted to kidney transplant with registered live donor organs and illegally procured organs) and was thus included twice in the analysis.After data extraction, another review was carried out on the selected papers to make sure the data for each individual was included only once in the analysis.
The estimated global prevalence of TB was calculated using a fixed and a random model.When heterogeneity was statistically significant, the random model was used.Meta-regression was used to identify possible sources of heterogeneity between studies.
Initially, univariate analysis was performed and all variables associated with TB prevalence rates in individuals submitted to kidney transplantation (p ≤ 0.2) in univariate analysis were included in the final multivariate meta-regression model.Level of significance of was set at 5%.
Studies were divided into three groups based on the prevalence of TB in the general population of each country, as follows: high prevalence countries (> 40/100,000), medium prevalence countries (20-40/100,000), and low prevalence countries (< 20/100,000).4 The combined TB prevalence rates in individuals submitted to renal transplant was also determined for each group of countries.
References were managed with the aid of software EndNote X4 and statistical analyses were carried out on STATA 11.0 (Stata Corp, College Station, Tex.).
This study was carried out as per the recommendations of the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.11 There were no conflicts of interest from the authors of this study.Permits from the Research Ethics Committee were not needed for this study, as it used data from previously published papers.

results
A total of 253 papers published since 2000 were screened.Although the databases allowed searches for older papers, the authors opted to select papers published since 2000 because immunosuppressants still in use today were introduced then.After the initial assessment, 86 abstracts were analyzed, 63 full papers were read, and 41 were included in this systematic review, representing a population of 73,808 individuals (Figure 1).Ten other studies listed in the references of the selected papers were analyzed, but none was included in the review.
Chart 1 lists the papers selected in this meta-analysis.
Table 1 summarizes the main characteristics of the studies included in this review.
The combined prevalence of TB among individuals submitted to renal transplant was 2.51% (95% CI = 2.17-2.85) in the random model and heterogeneity between studies was statistically significant (p ≤ 0.001).
Univariate analysis showed that the variables significantly correlated with differences in TB prevalence rates in individuals submitted to kidney transplants (p < 0.2) were year of publication, study design, sample size, use of secondary data source, and prevalence of TB in the general population.Therefore, these variables were included in the final meta--regression model, in which sample size > 2501 (p = 0.011) and high prevalence of TB in the general population (p < 0.001) sustained the association (Table 2).
Figure 2 shows the combined TB prevalence rates for individuals submitted to renal transplants based on the categorization of countries by prevalence of TB.This figure shows that in low prevalence countries the TB prevalence rate was 0.56% (95% CI = 0.40-0.73), in medium prevalence countries it was 2.61% (95% CI = 1.75-3.46),and in high prevalence countries the rate was 6.88% (95% CI = 5.11-8.65).

dIscussIon
This systematic review included observational studies on the prevalence of TB in individuals submitted to kidney transplants.Papers from different geographic areas were identified, including countries with high, medium, and low TB prevalence rates.Despite the greater number of studies conducted in high prevalence countries, possibly due to the higher risks of disease dissemination, there has been equal concern with describing the prevalence of TB in individuals submitted to renal transplants in countries with low and medium TB prevalence. 10, The mbined prevalence of TB in individuals submitted to kidney transplant of 2.51% (95% CI = 2.17-2.85) is 14 times greater than the prevalence rate of 0.18% (95% CI = 0.16-0.20) 4 seen in the general population.However, the analysis of countries with high TB prevalence rates revealed that the prevalence of TB in kidney transplant patients was 43 times greater than in the general population (6.88% vs. 0.16%, 4 respectively); in medium prevalence countries, the prevalence of TB in transplant patients was 83 times greater than in the general population (  respectively); in low prevalence countries, transplant patients had TB prevalence rates 56 greater than the general population (0.56% vs. 0.01%, 4 respectively).These findings were in agreement with previous population studies, but no papers were found to have carried out such global analysis. 7,8,52 meta-regression analysis, the prevalence of TB in individuals submitted to renal transplant was correlated with high TB prevalence rates in the general population (p < 0.001).Countries with high TB prevalence rates were found to have higher rates of infected individuals and individuals with disease.This correlation was expected, as transplant patients are given immunosuppressants and are thus at increased risk of infection by Mycobacterium tuberculosis. 5,36,53Yet, the exponential growth of prevalences seen in transplant patients against the general population indicates management strategies must be tailored to specifically address the needs of each population group. 6owever, the analysis of findings must consider the limitations inherent to this study.An important factor -TB prophylactic care -was not included in the meta-regression analysis, as only a few studies looked into this parameter. 13,26,35,37,42Another limitation pertains to the adopted method -database search -which may have left out non-indexed, yet possibly eligible studies.
Despite the study's limitations, significant variability was seen in the prevalence rates of TB among individuals submitted to renal transplant, and there appears to be a strong correlation between general population and kidney transplant patient population TB prevalence rates.Additionally, the lack of a specific protocol to assess the quality of the studies was overcome by the stratification of the analyzed variables and the adopted regression model, in a way not to affect our results.
The growing number of transplants pushed by the chronic kidney disease pandemic, and consequently of individuals on immunosuppressants at a greater risk of contracting diseases such as TB, [54][55][56] combined with the less than effective strategies to reduce morbimortality from TB,6 call for the immediate planning of preventive efforts and specific disease control measures, so that the already high prevalence rates seen in this population do not grow even further in the coming years.

Figure 2 .
Figure 2. Combined prevalence of tuberculosis in renal transplant patients according to country categorization per levels of TB prevalence in the general population. references 4.61% vs. 0.03%,4

tAble 2
cOrrelatiOn between study variables and estimated prevalence Of tuberculOsis in individuals submitted tO kidney transplant