GENETICS has revolutionized medicine. The human genome project has succeeded in sequencing almost all 3 billion nucleotides present in the human genome, an accomplishment that has been hailed as one of the greatest achievements of our time. Although more than 99% of human DNA is identical between individuals, residual variability not only makes each person unique but also, in the context of medicine, may contribute to disease onset or disease progression. As a result, the next step in the human genome project is designed to investigate DNA variability between species as well as between humans. Harnessing genetic information from clinical studies to examine the impact of genetic variability on disease characterization and outcome is called functional genomics. In this issue of Anesthesiology, the study by Lasocki et al. 1provides an example of how genetic variation in a common gene (the angiotensin converting enzyme [ACE] gene) alters pressure and flow relations during cardiopulmonary bypass. In this study, the authors use the unique environment of the operating room to answer a potentially important and clinically relevant physiologic question and, in the process, give new meaning to the operating room as the “last human physiology laboratory in medicine.”
Anesthesiologists have long recognized that response to drug administration or stress depends on the individual. In fact, a bell-shaped curve of responses to various environmental perturbations (drug administration, hemodynamic challenge, inflammatory response to stress of surgery, and others) shows that although most patients respond in predictable patterns, others respond either more or less vigorously. In fact, much of the art of anesthesiology is the astute clinician being prepared to deal with outliers. Increasingly clinicians are appreciating that an individual patient's response to stress may alter perioperative outcomes such as incidence of respiratory distress syndrome, perioperative myocardial infarction, survival, and response to pain management—but what are the mechanisms underlying pharmacodynamic and physiologic variability to stress? The answer to this complex question includes understanding how the unique genetic background an individual brings to the operating room affects his or her perioperative outcome.
To understand genetic variability in patients, it is important to briefly review some basics of genetics. The human genome consists of the aggregate of DNA present in 23 chromosomes (22 chromosome pairs plus sex chromosomes X/Y). DNA consists of a collection of nucleotides (A, C, G, or T) that form a code used to produce proteins. The expression of genetic information in the cell is generally a one-way system. In simple terms, the sequence of nucleotides in DNA determines the sequence of RNA via transcription, and then a sequence of three nucleotides of RNA encodes one amino acid (out of 20 possible amino acids) via translation; these amino acids are the building blocks for proteins, which perform virtually all chemical reactions and functions in the body. Only a small proportion of the entire DNA in a cell actually codes for RNA and proteins. Some of the nontranscribed DNA has controlling functions over transcription; thus, certain DNA nucleotide sequences may silence or enhance transcription. The word polymorphism refers to an altered DNA sequence compared with the most common, or wild-type, DNA. Polymorphisms can either be small or large insertions or deletions of nucleotides, repetitive sequences (often repeated pairs of DNA called microsatellites), or a change in a single nucleotide (single-nucleotide polymorphism). Some genetic variation provides us with unique (but nonmedically important) aspects of our bodies, such as facial expression, hair color, and body stature, but other variations have been shown to be important in determining onset and severity of disease, as well as response to drug therapy. Polymorphisms may directly alter the amino acid sequence of a protein and therefore potentially alter protein function, but other variants may alter regulatory sequences in DNA, thereby altering concentrations of otherwise normal proteins. Distinguishing background genetic variation that make us unique and provides a genetic fingerprint, from clinically important polymorphisms that lead to disease is one of the most important tasks for clinical researchers today. Classically, pharmacologists have concentrated on genetic variability that alters drug metabolizing enzymes to explain variation in pharmacokinetic responses to drug therapy; however, it is now apparent that the genetic variability can alter many other proteins important to pharmacodynamic responses. In the operating room, clinical research needs to go one step further and examine how genetic variability affects responses to acute stress as well.
Historically, human genetic studies have focused on relatively rare Mendelian-inherited, single-gene diseases. Diseases such as sickle cell anemia are the result of a single-nucleotide polymorphism in the DNA sequence encoding the β chain of adult human hemoglobin; the disease is entirely the result of this single-nucleotide polymorphism and thus is relatively easy to investigate. In less-well-characterized inherited diseases, genetic studies often focus on linkage of a genetic locus (region of DNA within a chromosome) with patterns of inheritance within families. In contrast to single-gene disorders, most common diseases prevalent today involve a complex interaction between a number of disease-enhancing or -altering genes, environmental stimuli, and variable (and sometimes age-dependent) penetration of those genes. 2Diseases such as hypertension, coronary artery disease, and some forms of cancer are examples of complex common diseases. Association studies analyze the incidence of a disease trait (or phenotype) and its association with a particular DNA polymorphism. Specifically, association studies test whether a genetic marker (or polymorphism) occurs more frequently in cases than in controls. The study by Lasocki et al. 1in this issue of Anesthesiology is an association study focusing on a candidate gene (the ACE gene) in which a well-described biologic effect occurs as a result of an insertion/deletion (I/D) polymorphism; this polymorphism impacts serum and renal concentrations of ACE. The authors are to be commended on using cardiopulmonary bypass as a unique and reproducible physiologic environment where blood flow can be altered and resulting arterial pressure (and systemic vascular resistance) can be documented, giving a measure of pressure–flow relations. Genetic association studies are only as good as the clinical characterization of end points (or phenotyping). The authors are therefore careful to include many factors that might affect systemic vascular resistance on cardiopulmonary bypass (e.g. , age, sex, hypertension history, medications, type of surgery) to ensure their results occur because of the ACE polymorphism and not some other clinical variable; however, other clinical covariates may also be important, such as body mass index, race, and central venous filling (which could be controlled by ensuring similar central venous pressure before beginning the study in all patients). Despite these limitations, the authors seem to demonstrate in a new setting that the specific ACE polymorphism tested is associated with permutations in pressure–flow relations in humans undergoing surgery.
Because association studies are difficult to perform well, especially in patient populations where genetic admixture is an issue (patients originating from many distinct genetic backgrounds), it is important to take this opportunity to stress some ideal standards for future genetic association studies. 3First, it is important to study a large sample population when testing for association with a genetic marker or polymorphism and disease. 4Citizens from Western countries originate from all over the world, bringing unique and varied genetic backgrounds that may have nothing to do with disease. In simplistic terms, this introduces genetic “wobble” that can only be overcome by either properly powering a study for a primary main end point with thousands of patients, or by carefully screening the populations to be as homogeneous as possible. Second, genetic association studies are only as good as the clinical phenotyping. Clinical end points must be quantifiable and reproducible. Clinical covariates that might explain different outcomes must be taken into account (e.g. , the effect of body mass index in a study designed to examine blood pressure may be far more important than variation in DNA sequence of a candidate gene). Third, use of surrogate markers or intermediate end points often augment a genetic association study. In the study in this issue of Anesthesiology, measurement of serum ACE concentrations might have validated the authors’ final conclusions. Fourth, rigorous quality control must be in place to define the presence of a genetic polymorphism. There are currently many laboratory methods available for polymorphism detection. On careful examination, a single-nucleotide polymorphism that occurs in 2 patients out of 100 might be missed or overcounted if the method of detection is only 92–96% accurate. Fifth, smart genetics must be used; a genetics collaborator can often be enlisted for this aspect of the trial. The use of appropriate and accurate genetic language is important so that concepts and results are universally recognizable. For example, it is more informative to list allele (or marker) frequencies (range, 0–0.50) rather than genotype frequencies (often listed as 0–100%) when describing variants. Race has been shown to be an important determinant of allele frequency in many studies, so allele frequencies for the overall group and then for each racial subgroup should be examined. If not appropriately taken into account, varying allele frequencies in subgroups might significantly influence final results of an association test. It is also important to test genetic markers to see if they are in Hardy-Weinberg equilibrium so that statistical genetics tools that require this condition are appropriately applied. Sixth, appropriate statistical analysis must be applied. Seventh and last, genotypes exist either as homozygous or heterozygous. Heterozygous genotypes may present as intermediate phenotype and therefore cannot necessarily be grouped together with homozygous genotype for the purpose of statistical analysis. A common mistake in association studies is to perform repeated comparisons without altering the final P value required for significance. As a general approach, overall significance should be sought, followed by analysis of covariants, determination of specific P values, and possible interaction between genes or polymorphisms. Appropriately performed analysis for association studies may be complicated, so statistical genetics expertise should be sought.
In summary, the study by Lasocki et al. 1provides a first step in demonstrating the usefulness of association studies in the anesthesiology literature. Association studies can be useful in determining the pathogenesis of disease, variability in physiologic end points, or response to treatment. If one is visionary, such studies may hold the key to unlock the secret of predicting perioperative outcomes based on individual preoperative genetic information. As such, they have the potential to revolutionize clinical research.