Acute renal injury (ARI) is a serious and frequent condition in critically ill patients. Its clinical expression varies greatly from elevations in serum creatinine concentration to anuria and the need for renal replacement therapy (RRT).1 The incidence of ARI among hospitalized patients is approximately 7%,2 but reaches 40% in critical patients with sepsis.3 In Spain, Herrera-Gutierrez et al.,4 in the context of a multicenter study conducted in 1999 (FRAMI), found the incidence of ARI upon admission to the Intensive Care Unit (ICU) to be 5.7%.

The host response to environmental stimuli is specific of each individual patient, and is conditioned by the underlying genetic profile (genotype). Two non-consanguineous individuals share 99.9% of their nucleotide sequence. However, the relatively small 0.1% differentiating portion suffices to determine individual character within one same species.5 The basis of this variability resides in genetic polymorphisms, which are variants in the DNA chain in a specific location (locus) that determine the existence of at least two different alleles.6 The most frequent allele is referred to as the native allele, while the rest are polymorphic alleles. These variants can comprise: (a) a single nucleotide, conforming a single nucleotide polymorphism (SNP); (b) a sequence, in which case the polymorphism is defined by the insertion or deletion of a fragment; or (c) variants in the number of repetitions of a given nucleotide sequence.6–9

Polymorphisms exert their influence: (a) directly, through the modification of risk mediated by the polymorphism; (b) as a “linked” group or haplotype, through the modification of risk mediated by a set of polymorphisms located in regions close to the DNA sequence, and which therefore do not combine; or (c) as a “non-linked” group or combination of polymorphisms, through the modification of risk mediated by genetic variants that are not inherited jointly and are not close to each other in the DNA chain, but which determine a concrete risk if they coincide within the genome of one same individual.10

The aim of the present systematic review is to analyze the existing evidence on the influence of genetic polymorphisms upon the risk of developing ARI and its outcome.

The review included articles published in English or Spanish analyzing the association between polymorphisms and: (a) susceptibility to ARI in patients versus healthy individuals, and between different groups of patients; (b) the severity of ARI. We excluded studies published only in abstract form, clinical cases, or studies including patients under 16 years of age, on chronic dialysis, or subjected to renal transplantation.

A total of 39 studies were identified; 27 were excluded and 12 were analyzed (Fig. 1). The polymorphisms associated to ARI susceptibility or outcome are presented in Table 1.

Figure 1.

Study selection flowchart.

(0.1MB).

The present review has found that polymorphisms of 11 different genes are associated to acute renal injury (ARI) susceptibility or to the severity or prognosis of ARI in different populations, while four of the analyzed genes showed no such association. The review comprised 4835 patients, with a mean of 402 participants per study. Only two investigations16,22 involved over 900 patients. The reproducibility of the analyzed polymorphisms is scarce, being limited to the APO E gene,19,21,22 ACE I/D with its AT1 receptor,21,22 and ACE isolatedly.11,12 The results are difficult to interpret, since the inclusion criteria, the end events observed, the sample sizes and the statistical analyses differed from one study to another. As an example, two of the three studies that included APO E concluded that the alleles different from ¿4 are associated to a poorer outcome. The third article, involving an almost four times larger sample size, recorded no association. In a similar way, of the four studies that explored polymorphism ACE I/D, one found genotype II to be associated with increased risk,11 another found allele D to be associated with increased risk,21 and two studies reported no association.12,22

The main potential applications of genetic analysis in clinical practice are:

• a.

  Risk identification and quantification. The application of genomic information to estimate the risk of developing ARI is currently not a part of routine practice despite the fact that the strength of the association (e.g., evaluated by means of the odds ratio (OR)) of the genetic and clinical factors is relatively similar. As an example, the OR range for the genetic factors varies between 0.58 and 6.50, depending on the polymorphism and event analyzed. On the other hand, in a recent study that included 1345 patients who were subjected to heart surgery,23 it was seen that the four clinical factors independently associated to ARI were: (i) anesthetic risk category 3 or 4 of the American Association of Anesthesiology (ASA score 3–4)(OR 2.60 [1.03–6.55]); (ii) forced expiratory volume in the first second (FEV1)(OR 0.55 [0.32–0.96]); (iii) the need for vasopressor drugs (OR 1.02 [1.0–1.03]); and (iv) the duration of anesthesia (OR 1.04 [1.00–1.08]).

• b.

  Generation of new etiological and physiopathological knowledge. Identification of the polymorphisms associated to ARI and of the way in which they influence expression of the respective genes will increase our knowledge of the etiology and physiopathology of ARI.

• c.

  Gene therapy. While still in the experimental phase, a number of procedures have already been evaluated, and designed to insert functional copies of genes that are defective or missing in the genome of a given individual.24 An example of this is transfection of the Bcl-2 gene in the rat kidney, with the aim of reducing ischemia-reperfusion damage through modulation of the apoptotic and necrotic pathways.25

• d.

  Pharmacogenetics. The analysis of the variability of drug response according to the genotype of the individual can result in greater therapeutic benefits and a minimization of adverse effects.10 As an example, the last step in the formation of adrenalin from noradrenalin is mediated by the PNMT enzyme. Alam et al.16 found that allele A of SNP −161 PNMT is associated to lesser mortality, and that AA homozygotes present lower urine adrenalin levels and greater noradrenalin levels. It therefore could be postulated that carriers of polymorphism SNP −161AA of the PNMT gene would benefit from the use of non-catecholaminergic vasopressors instead of catecholaminergic agents. SNP COMT +472 G/A, which determines substitution of the amino acid valine by methionine in codon 158, reduces the concentration of the enzyme and favors its transformation into a thermolabile form.22,26 Since COMT is essential for catecholamine degradation in the kidney, the finding by Haase-Fielitz et al.20 that genotype AA is associated to an increased risk of ARI and shock could be related to the catecholamine desensitization phenomenon. This interpretation is supported by the fact that the serum concentrations of adrenalin and noradrenalin 6h after surgery were greater. The described genetic association and the suggested physiopathological hypothesis, if confirmed, would also be relevant, since patients with genotypes of this kind could represent another group capable of deriving benefit from the use of non-catecholaminergic vasopressors.

Despite the potential applications of this new genetic knowledge, the available information is still partial and difficult to interpret. As an illustrating example, oxidative stress is an important etiopathogenic factor in ARI. SNP p22phox +242C/T, which determines substitution of the amino acid histidine by tyrosine in position 72 of the polypeptide, modulates binding affinity to the heme group, and thus the activity of the enzyme.27–29 Allele T is associated to lesser respiratory activity in endothelial cells28 and in human neutrophils.29 However, the studies that have analyzed the influence of SNP +242 upon the lipid peroxidation markers have yielded conflicting results. Nitrotyrosine is the product of tyrosine nitration by reactive nitrogen species such as the peroxynitrite anion and nitrogen dioxide; as a result, it is regarded as a marker of oxidative stress dependent upon nitric oxide. Paradoxically, in the study published by Perianayagam et al.,14 the NADPH oxidase p22phox +242 TT genotype was associated to increased levels of nitrotyrosine. As a possible explanation for this, the investigators suggested the existence of linkage between this SNP and another still unknown polymorphism, which would be the true element responsible for exerting the genetic influence.

A second example of the difficulties found in the interpretation of the results is represented by SNP HIF 1α +1772. Hypoxia plays a role in the development of ARI. HIF 1α is activated in situations of oxygen deficiency and activates a series of genes that are responsible for increasing cell resistance to hypoxia.30 SNP HIF 1α +1772C/T, which determines substitution of the amino acid proline by serine in position 582 of the protein, is associated to an increment in the transcription of the gene. The finding by Kolyada et al.15 is unexpected, since a priori the increase in the expression of this group of genes should increase resistance to hypoxia. Possibly, increased expression of the gene generates an excessive response that could persist over time, even after resolution of the ischemia, and thus ultimately would turn into an adverse or maladaptive response.

Classical research focuses on identification of the components of biological systems.31 The tremendous pace of technological progress has allowed a shift in the paradigm toward a holistic approach in which complex biomolecular networks are considered globally, attempting to explain diseases or phenotypes on the basis of interactions among the host genome, genic expression and the environment—in consistency with a systems biology approach. The exploration of candidate genes was the strategy used in all the publications considered in our review. This methodology, of great strength when it comes to interpreting the results, involves analysis of the associations among polymorphisms of genes integrating the physiopathological pathways with observable characteristics. However, the studies are limited to relatively few polymorphisms. Studies that investigate the entire genome (genome-wide association studies, GWAS) allow global exploration of the variants included in the nucleotide sequence, and have been successful in identifying associations with many complex illnesses.32 This strategy has not yet been applied to ARI, however.

In recent years, the challenges found in genetic and genomic studies have increased. In ARI, knowledge would be required of genomic and exomic sequences, along with their functional characterization; of the order and magnitude of the activation/inhibition of certain genes; and of their interaction with the epigenetic regulation levels and with the environment.33–35

The genetic structure of the patient influences susceptibility to ARI and the severity and outcome of the condition once established. Such new knowledge could be applied in clinical practice for: (a) risk identification and quantification; (b) the generation of new etiological and physiopathological knowledge; (c) the design of new treatments based on gene therapy; and (d) the prediction of individual patient response to certain drugs.

At present, all the published studies use an approach based on individual candidate genes, which substantially limits interpretation of the results. In future, studies capable of analyzing thousands of polymorphisms of the entire genome (GWAS) in a large number of people, of different ages, different ethnic groups and different geographical areas could help expand the existing knowledge.

The investigation that has produced these results received financial support from the Seventh Framework Program of the European Union PM7/2010, by virtue of funding agreement no. 26486, and the collaboration of the Instituto de Salud Carlos III (FIS 08/1726), the Fundación Mutua Madrileña (AP/67842009) and the Lilly Foundation-Spain.

The authors declare that they have no conflicts of interest.