Gender in the allocation of organs in kidney transplants: meta-analysis

OBJECTIVE To analyze whether gender influence survival results of kidney transplant grafts and patients. METHODS Systematic review with meta-analysis of cohort studies available on Medline (PubMed), LILACS, CENTRAL, and Embase databases, including manual searching and in the grey literature. The selection of studies and the collection of data were conducted twice by independent reviewers, and disagreements were settled by a third reviewer. Graft and patient survival rates were evaluated as effectiveness measurements. Meta-analysis was conducted with the Review Manager® 5.2 software, through the application of a random effects model. Recipient, donor, and donor-recipient gender comparisons were evaluated. RESULTS : Twenty-nine studies involving 765,753 patients were included. Regarding graft survival, those from male donors were observed to have longer survival rates as compared to the ones from female donors, only regarding a 10-year follow-up period. Comparison between recipient genders was not found to have significant differences on any evaluated follow-up periods. In the evaluation between donor-recipient genders, male donor-male recipient transplants were favored in a statistically significant way. No statistically significant differences were observed in regards to patient survival for gender comparisons in all follow-up periods evaluated. CONCLUSIONS The quantitative analysis of the studies suggests that donor or recipient genders, when evaluated isolatedly, do not influence patient or graft survival rates. However, the combination between donor-recipient genders may be a determining factor for graft survival.

Kidney transplants are considered to be the best therapeutic alternatives for persons suffering from advanced chronic kidney disease. 13,18,32 Gender differences regarding kidney transplants have been reported in the literature and observed in the clinical practice over the last decades. They affect transplant results, such as in acute and chronic rejections and graft and patient survival rates. Women have less access to transplants. They have increased risk of acute rejection and decreased risk of chronic rejection -those risks increase with age. 9,22 In turn, women are observed to account for around 65.0% of living kidney donors. 7,37 The etiology of those differences is still unknown, but it probably reflects hormone, immunological, and aging differences, as well as prejudice. 7,22,28 Survival rates are higher among women following kidney transplants, 9,11 but the data are not confirmed by the literature. In a South African study, worse survival rates have been observed among women, but no significant differences were found between genders in graft survival. 23 In another study, no differences were observed between genders in patient and graft survival rates. 27 Knowing differences across genders is necessary to identify possible barriers in the achievement of ideal results and in the development of interventions that overcome those barriers. This review, by focusing on those gender-related differences in the clinical effectiveness of immunosuppressive therapies for kidney transplant maintenance, may promote better understanding, provide more efficient health care, contribute to the creation of clinical protocols, and promote better long-term results for patients.
The objective of this review was to analyze whether genders influence patient and graft survival rates in kidney transplants.

METHODS
This review was conducted according to the recommendations from the Cochrane Collaboration Handbook. a The article was prepared according to Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA). 21 Observational cohort studies were selected. The selection included studies with patients who received kidney transplants from living or deceased donors for the first time or more than once, mentioning gender differences concerning pre-transplant characteristics, and INTRODUCTION finding survival results for grafts, patients, or for both. Studies that did not involve immunosuppressants for maintenance of kidney transplants, pharmacokinetic studies, economic evaluation studies, review studies, and studies conducted on animals were excluded, as per the exclusion criteria. After duplicate studies were excluded, two independent reviewers selected the references in three phases: analysis of titles, abstracts, and full texts. The disagreements were settled by a third reviewer. The data -including methodological quality, subject information, treatment length, and patient and graft survival rates -were extracted and collected in duplicate in a Microsoft Excel 2010 spreadsheet.
Methodological quality evaluations were independently conducted by the reviewers, and related disagreements were settled through the consensus among reviewers. Newcastle-Ottawa scale b for observational studies was used. In this scale, each study is evaluated in three dimensions: selection of study groups; the comparability among groups; and the ascertainment of either the exposure or outcome of interest. Total score was up to nine stars -above six, studies are considered to be high quality.
In order to be analyzed, the studies were grouped according to the comparison of results among: The study data were combined using randomized effects model in Metaview module of Review Manager software, version 5.3. The results were presented as relative risk for dichotomous variables, with a confidence interval of 95%. Analyses with I 2 > 40.0% and p-value of Chi-squared test < 0.10 were considered to be significant heterogeneity. A sensitivity analysis was conducted to investigate heterogeneity causes, with the exclusion of one study at a time, with changes being verified in the values of I 2 and p.
The outcomes that were evaluated in the meta-analysis were graft survival and patient survival per follow-up period (one, two, three, five, eight, 10 years or more).

RESULTS
A total of 5,993 publications were initially identified in the electronic databases, and seven through manual searches, all adding up to 6,000 publications. Of these, 500 publications were excluded because of the participant type, 5,251 due to study type and 177 due to the intervention. The main causes for excluded studies were: studies that did not analyze the outcome of interest (patient and graft survival rates), ones that did not include kidney transplants, review, pharmacokinetic, pharmacoeconomic studies, among others. After duplicate publications were eliminated and the reviewers conducted their analyses, 29 cohort studies were included, which involved 765.753 patients. Among those, six studies compared the measurements from results involving donor genders (MD and FD); eight studies, involving recipient genders (MR and FR), and for 20 others, results involving donor-recipient Abstract)) OR (((((((((Factor, Sex) OR Factors, Sex) OR Sex Factor)) OR (Sex Factors)) OR ("Sex Factors"[Mesh)) OR (((((((Characteristic, Sex) OR Characteristics, Sex) OR Sex Characteristic) OR Sex Differences) OR Difference, Sex) OR Differences, Sex) OR Sex Difference)) OR (Sex Characteristics)) OR ("Sex Characteristics"[Mesh))) AND Humans[Mesh)) AND ( ("Transplantation, Kidney" OR "Kidney Transplantations" OR "Transplantations, Kidney" OR "Transplantation, Renal" OR "Renal Transplantation" OR "Renal Transplantations" OR "Transplantations, Renal" OR "Grafting, Kidney" OR "Kidney Grafting" OR "Kidney Transplantation" OR "MeSH descriptor:Kidney Transplantation 1 tree(s) exploded") AND ("male" OR "female" OR "gender" OR "Factor, Sex" OR "Factors, Sex" OR "Sex Factor" OR "Sex Factors" OR "MeSH descriptor:Sex Factors explode all trees") AND "Characteristic, Sex" OR "Characteristics, Sex" OR "Sex Characteristic" OR "Sex Differences" OR "Difference, Sex" OR "Differences, Sex" OR "Sex Difference" OR "Sex Characteristics" OR "MeSH descriptor:Sex Characteristics explode all trees") AND ("graft rejection OR MeSH descriptor:Graft Rejection explode all trees") AND ("survival rate" "MeSH descriptor:Survival Rate explode all trees") AND "graft survival" LILACS 87 (tw:(transplantation kidney)) OR (tw:(transplantation renal)) AND (tw:(gender differences)) OR (tw:(gender Characteristics)) OR (tw:(Sex Characteristics)) OR (tw:(sex differences)) AND (tw:(survival rate)) AND (tw:(survival graft)) genders (MD-MR, MD-FR, FD-FR, FD-MR) ( Figure).
One study was included in donor, recipient, and donor-recipient gender comparisons; 5 another one 14 was included in the comparison between recipient and donor-recipient genders; and two studies, in the comparison between donor and donor-recipient genders. 3,25 Out of the 29 observational studies included, 28 were retrospective and one was prospective. 38 Most studies have not reported average follow-up periods, and the data from the cohorts were collected from 1978 to 2009. In the comparison between donor genders, 9,673 subjects were evaluated in the six studies. In the comparison between recipient genders, 84,070 subjects were evaluated in the eight studies included. In the comparison between donor-recipient genders, 672,010 subjects were evaluated in the 20 studies included. In regards to types of donors, 12 studies evaluated deceased donors, eight evaluated living donors, and nine evaluated both kinds (living and deceased) ( Table 2).
The majority of subjects (donor, recipient, and donor-recipient) were males, for all gender comparisons. All studies included evaluated graft survival rates, but only eight of them evaluated patient survival rates. 1,14,20,22,23,27,39,43 Out of the 29 studies included, two of them were observed to have scores of six as per New-Castle Ottawa scale; 2,4 14 (48,3%) of them had score seven; and 13, eight ( Table 2).
Out of the six studies included in the systematic review for donor gender comparison, five of them were included in the meta-analysis for graft survival outcome. [3][4][5]29,41 The study by Neugarten et al 25 was not found to have enough numerical data for the quantitative analysis. The studies included in the donor gender comparison did not evaluate patient survival.
Regarding graft survival, the relative risks (RR), as grouped chronologically according to follow-up periods of one, two, three, five, and ten years were, respectively,  Out of the eight studies included in the systematic review for recipient gender comparison, six were included in the meta-analysis for graft survival outcome. 2,5,6,14,23,35 The studies by Nyberg et al 27 and Meier-Kriesche et al 22 were not found to have numerical data in order to be included in the quantitative analysis. Thus, meta-analyses were conducted for monitored periods of one, five, and 10 years or more.
In the meta-analysis regarding graft survival for one year, four comparisons of three studies were included. In the study by Sánchez Garcia et al, 35  . No significant differences were found for any of the follow-up periods, and no recipient genders were highlighted among the groups. Heterogeneity was high and significant for follow-up periods of five and 10 years or more (Table 3).
Regarding patient survival, two studies were included in the related meta-analysis. 14,24 The meta-analysis was conducted for the follow-up period of 10 years or more, heterogeneity was high and the difference was not significant (RR = 0.96; 95%CI 0.67;1.37; p = 0.81; I 2 = 95.0%) ( Table 3).  Table 4).
The remaining studies were not found to have enough data for the quantitative analysis. 3,10,12,14,25,42,45 For the patient survival outcome, two studies were included in the meta-analysis. 1,20 The studies by Ellison et al 8   follow-up period. The patient survival analysis was only conducted for the one-year follow-up period, and no pairs were observed to be favored (RR = 0.99; 95%CI 0.96;1.02; p = 0.52; I 2 = 0.0%).
In the MD-MR versus FD-FR comparison, the RRs as grouped for graft survival in a chronological order of follow-up periods of one, two, three, five, and 10 years were, respectively, 1. In the MD-FR versus FD-FR comparison, the RRs as grouped for graft survival in a chronological order of follow-up periods of one, two, three, five, and 10 years were, respectively, 1.01 (95%CI 1.00;1.03;

DISCUSSION
This meta-analysis evaluated the influences of donor genders, recipient genders, and the donor-recipient combination in regards to kidney transplant patient and graft survival rates.
In the comparison between donor genders, 9,022 subjects were evaluated in the meta-analysis. Donor genders were not found to favor gender rates in the evaluation of one, two, three, five, and eight-year follow-up periods (p < 0.05). Ten-year follow-up period was the only observed to differ significantly, favoring male donors (p = 0.03). The study by Muller 24 concluded that kidney grafts from male patients work better than the ones from female donors in the long run. Several studies suggest that, in those cases, the grafts from female donors are more antigenic, which may explain the lower survival rates. 22,30,31,36 In the comparison between recipient genders in all follow-up periods, 9,593 subjects were evaluated in the meta-analysis, and no significant differences were observed regarding graft survival. Patient survival analysis considered 747 patients. No significant differences were found among the studies. The study by Busson and Benoit 5 evaluated the influence from recipient and donor genders and the donor-recipient combination. Recipient genders were the only ones for which significant differences were not found. The assessment of differences between genders is important to improve transplant results. Men and women have different biological factors, different body conditions, hormone circumstances, and immune responses, as well as different metabolic and functional demands, which can influence kidney transplant results. 44 Gender incompatibilities in FD-MR pair are argued to negatively influence graft survival due to kidney sizes and their numbers of nephrons. The ratio between graft and recipient weights is an important one. 26,33 Giral et al 10 analyzed the consequences from kidney mass reduction following kidney transplants, and they concluded kidney graft mass to impact glomerular filtration and proteinuria rates. The authors suggest that great kidney-to-recipient weight ratios be avoided, once that might significantly influence long-term kidney function.
That meta-analysis only included cohort studies. One of the limitations from systematic reviews with meta-analyses of observational studies regards to the selection bias that is intrinsic to this study design and to uncontrolled confounding factors. Observational cohort studies are the ones conducted in the real world, under conditions uncontrolled for. In regards to that, differences were observed in the subject numbers among the groups, types of donors (living, deceased, or both), numbers of transplants, monitored periods, among others. Despite that, observational studies are observed to have the advantages of gathering a large number of patients and best representing the real world.
Another limitation in the interpretation of results was the statistical heterogeneity among studies, which was found in the meta-analyses. The small number of studies included in the comparisons, and the lack of complete, accurate information in the studies made it difficult to account for heterogeneity sources. Most studies were not observed to include immunosuppressive therapies, ages (donors and recipients), or monitored periods. The sensitivity analysis, in which studies were included and excluded for each comparison, in general, has not altered the directions of outcomes, having changes of small relevance in heterogeneity values. It has not provided information on the possible causes for heterogeneity either.
However, the results from this review can be considered for decision-making by medical teams responsible for transplants in the clinical practice, once it represents the best level of evidence regarding the topic.
Gender incompatibilities must be avoided whenever possible. Genders must be considered as criteria in the choices regarding allocation of organs from donors and to recipients. Gender combinations may make a difference in survival rates. Nonetheless, that reality is utopic in the clinical practice, due to the scarcity of donors and the increase in the number of patients on waiting lists for transplants. A change in that scenario is required in order to improve the allocation of organs.
In conclusion, recipient and donor genders, when evaluated isolatedly, do not influence patient or graft survival rates. However, combinations between donor-recipient genders may be a determining factor for graft survival, favoring MD-MR pair.