Clinical Abstracts
| Clinical Abstracts |
|
|
|
August 2008 Editor:
Hyperhomocysteinemia and myocardial expression of BNP Hyperhomocysteinemia and chronic heart failure frequently are problems for the elderly and often coincide. Epidemiologic studies from Europe and the United States consistently have found that the frequency of hyperhomocysteinemia (HHcy) is more than 25 percent in persons 55 years old and older and increases continuously with age. Chronic heart failure (CHF) affects 5 million patients in the United States, and approximately 550,000 cases are diagnosed each year. Hospital discharges for CHF increased from 399,000 in 1979 to 1,093,000 in 2003, and the 2003 overall death rate for CHF was 19.7 percent. A similar situation exists in Europe. Consequently, it is important to prevent CHF by identifying and modifying risk factors. Previous analyses found hypertension, smoking, diabetes mellitus, obesity, and advancing age to be the most important risk factors for CHF. Recent clinical and experimental studies suggest that HHcy is also a risk factor for CHF. Previous studies from the authors’ group, analyzing more than 1,000 people, found a consistent link between total plasma homocysteine (tHcy) and severity of CHF. The closest relation could be observed between tHcy and N-terminal pro-brain natriuretic peptide (NT-proBNP). In this study, the authors investigated myocardial BNP expression and cardiac remodeling in healthy adult rats after three months of moderate or intermediate HHcy. The authors randomized 50 rats into five groups. Groups Co1 and Co2 (controls) received a typical diet. Groups Meth, Hcy1, and Hcy2 were fed the same diet supplemented with 2.4 percent methionine, one percent homocystine, and two percent homocystine, respectively. After 12 weeks, the authors measured tHcy and BNP in plasma and tissue and performed histomorphometric analyses. The authors found that all of the animals had comparable baseline body weights (mean [SD] 234 [26] g) and levels of total circulating Hcy (4.7 [1.7] µmol/L). After 12 weeks of treatment, total circulating Hcy increased in Meth, Hcy1, and Hcy2 (27.3 [8.8], 40.6 [7.0], and 54 [46] µmol/L, respectively) and remained unchanged in Co1 and Co2. Serum BNP significantly increased in one of 10 animals in Meth, three of 10 in Hcy1, and three of 10 in Hcy2. Median (25th–75th percentile) BNP tissue concentrations in Hcy1 and Hcy2 were 55 percent higher than in the corresponding controls (Co1 versus Hcy1, 225 [186–263] versus 338 [262–410] pg/mg protein, P=.05; Co2 versus Hcy2, 179 [107–261] versus 308 [192–429] pg/mg protein, P=.12). In the Meth group, BNP expression was comparable to that of controls (200 [159–235] versus 225 [186–263] pg/mg protein, P=.32). The percentage of perivascular and interstitial collagen and mast cell infiltration were comparable in all groups, indicating no adverse cardiac remodeling. The authors concluded that three months of intermediate HHcy stimulated increased cardiac BNP expression that was not accompanied by adverse cardiac remodeling. Herrmann M, Taban-Shoma O, Hubner U, et al. Hyperhomocysteinemia and myocardial expression of brain natriuretic peptide in rats. Clin Chem. 2007;53:773–780. Reprints: Wolfgang Herrmann, Abteilung für Klinische Chemie und Laboratoriumsmedizin/Zentrallabor, Universitatsklinikum des Saarland, D-66421 Homburg/Saar, Germany; kchwher@uniklinik-saarland.de
Preeclampsia affects approximately five percent of all pregnancies, and its causes remain unclear. A state of insulin resistance has been demonstrated in active preeclampsia, and women with clinical evidence of insulin resistance are at higher risk of developing this syndrome during pregnancy. A family of putative insulin mediators, called inositol phosphoglycans (IPGs), recently have been implicated in the pathophysiology of preeclampsia. A high concentration of bioactive IPG P-type (P-IPG) was found in human preeclamptic placenta, urine, and amniotic fluid. Additional studies have provided evidence of insulin resistance in human placentas from preeclamptic pregnancies and demonstrated an association between P-IPG and the insulin signaling cascade. IPGs have been shown to exert insulin-mimetic activity in glucose and lipid metabolism, primarily by activating pyruvate dehydrogenase phosphatase, glycogen synthase phosphatase, and glycerol-3-phosphate acyltransferase. The elevated concentration of glycogen in terminal villi from preeclamptic placentas, due to increased activity of glycogen synthase, may be a consequence of the abnormal content of P-IPG in the placentas of preeclamptic patients. The authors conducted a case-control study to assess the potential of urinary levels of P-IPG as a screening test for preeclampsia. They recruited for the study 27 preeclamptic women and 47 healthy pregnant women. A polyclonal antibody-based enzyme-linked immunosorbent assay was developed to detect levels of P-IPG in urine. Levels of P-IPG in urinary specimens were found to be 30-fold higher in preeclamptic subjects than in control subjects (329.1±21.8 versus 9.2±1.5; P<.001), with a higher level in all of the preeclamptic cases. For six women who developed preeclampsia, more than one gestational date sample of urine was available, and retrospective analysis showed a significant time-related increase of the urinary level of P-IPG seven weeks or less before clinical diagnosis of preeclampsia. Urinary level of P-IPG increased after diagnosis, indicating a possible pathophysiological threshold level, and steeply decreased after delivery. Williams PJ, Gumaa K, Scioscia M, et al. Inositol phosphoglycan P-type in preeclampsia: a novel marker? Hypertension. 2007;49:84–89. Reprints: Marco Scioscia, Dept. of Obstetrics and Gynecology, University of Bari-Policlinico, Piazza Giulio Cesare 11, 70125 Bari, Italy; marcoscioscia@gmail.com
Septic reactions continue to be reported with culture-tested platelet products, likely due to false-negative results associated with inadequate sampling of low-concentration bacteria. The commercially available bacterial detection system BacT/Alert 3D (BioMérieux, Hazelwood, Mo.) is cleared by the FDA for quality control testing of apheresis and whole blood-derived platelets. The assay is validated to detect common contaminants at concentrations of 1 to 10 colony-forming units (CFUs) per mL (300–3,000 CFUs in a 300-mL product), yet platelet products may be contaminated on collection at levels as low as 1 CFU per product (0.003 CFU/mL in a 300-mL product). In practice, a 24- to 36-hour delay between collection and sampling is employed on the assumption that bacteria will move from the lag phase into a log phase of growth, and low initial inocula will grow to exceed the levels necessary for detection. A number of reports document that culture systems reliably detect bacterial contamination in apheresis platelets with a range of organisms at a frequency of 1 in 5,000 to 1 in 6,000 and prevent transfusion of culture-positive units in a majority of cases. Despite these successes, continued reports of septic transfusion reactions and fatalities after the transfusion of culture-negative platelets raise the possibility that the BacT/Alert system has an appreciable failure rate. In two-bottle (aerobic and anaerobic) culture systems, the frequency of growth in one bottle or in both culture bottles is a measure of bacterial concentration at the time of sampling for organisms that can grow in both conditions. Published reports reveal a mean residual risk of sepsis of 2.3 per 105 products tested. Modeling the effect of concentration predicts that 50 and 95 percent of samples are detected at 0.09 and 0.36 CFU per mL for organisms that grow under aerobic and anaerobic conditions and suggests a 50 and five percent false-negative culture rate at these concentrations. Re-analysis of published studies documenting single-bottle growth of nonfastidious organisms indicates that low bacterial concentrations are frequently encountered and predict false-negative cultures in products contaminated with common bacteria at rates that varied from four percent to more than 75 percent. The model highlights the weakness of the American Association of Blood Banks’ definition of false-positive culture results at low bacterial concentrations, especially for organisms that grow poorly in aerobic storage conditions. The authors concluded that continuing reports of septic reactions after transfusion of culture-tested apheresis platelets and frequent single-bottle-positive results when testing with two-bottle systems suggest appreciable levels of false-negative results with BacT/Alert as implemented. Benjamin RJ, Wagner SJ. The residual risk of sepsis: modeling the effect of concentration on bacterial detection in two-bottle culture systems and an estimation of false-negative culture rates. Transfusion. 2007;47:1381–1389. Reprints: Dr. Richard J. Benjamin, Biomedical Services, National Headquarters, American Red Cross, 2025 E St., NW, Washington, DC 20006; benjaminr@usa.redcross.org
Amino-terminal pro-brain natriuretic peptide (NT-proBNP) is an important risk marker in cardiovascular disease. Release of brain natriuretic peptides is caused by wall stress of the atria and ventricles. However, plasma peptide concentrations have also been shown to be associated with a number of demographic and clinical variables. In most cases, the mechanism behind these associations is poorly understood. It is well known that concentrations of natriuretic peptides increase with age, and while this finding does not appear to contradict intuition, a clear mechanism has not been elucidated. The association between age and NT-proBNP may be due to increased release or decreased clearance of NT-proBNP with increasing age, but further investigation is needed. The apparent paradox that NT-proBNP decreases with increasing body mass index (BMI) is more complex. The mechanisms behind the association between BMI and NT-proBNP are unclear. An association between renal function and plasma concentrations of NT-proBNP has been documented several times and may be caused by small changes in renal extraction of NT-proBNP because approximately 15 to 20 percent of the NT-proBNP delivered to the kidneys is extracted. Obesity is a state characterized by glomerular hyperfiltration and age-related decreases in glomerular filtration rate (GFR). BMI, age, and GFR are associated with plasma concentrations of NT-proBNP in chronic heart failure patients. The authors tested the hypothesis that the effects of BMI and age on plasma concentrations of NT-proBNP are associated with GFR. They obtained clinical data and laboratory test results from 345 chronic heart failure patients at a baseline visit to the authors’ heart failure clinic and examined the hypothesis using multiple linear regression models. Age (P=.0184), BMI (P=.0098), hemoglobin (P=.0043), heart rhythm (P<.0001), and left ventricular ejection fraction (P<.0001) were associated with log(NT-proBNP). After adjusting for GFR estimated by the Cockcroft and Gault equation, the parameter estimates for BMI (P=.3807) and age (P=.7238) changed markedly and became insignificant. In another model, after adjusting for GFR estimated by the four-component Modification of Diet in Renal Disease formula (eGFRMDRD), the parameter estimates for age (P=.0674) changed markedly and became insignificant, but BMI (P=.0067) remained significant and unchanged. The eGFRMDRD is adjusted for body surface area, which may explain the difference. In chronic heart failure patients, the effect of age on NT-proBNP is associated with estimates for GFR derived from serum creatinine, and the significance of the effects of BMI on NT-proBNP depends on the method by which GFR is estimated. Schou M, Gustafsson F, Kistorp CN, et al. Effects of body mass index and age on N-terminal pro-brain natriuretic peptide are associated with glomerular filtration rate in chronic heart failure patients. Clin Chem. 2007;53:1928–1935. Reprints: Dr. Morten Schou, Dept. of Cardiology and Endocrinology, Clinic E, Frederiksberg University Hospital, Ndr. Fasanvej 57-59, DK-2000 Frederiksberg, Denmark; m.schou@dadlnet.dk
Interferon regulatory factor 5 (IRF5) is involved in the production of cytokines implicated in the pathophysiology of rheumatoid arthritis, such as tumor necrosis factor α, interleukin 6, and type I interferon. The haplotype bearing the IRF5 rs2004640-T allele (XTXT), which confers increased expression and unique splicing variants, recently was shown to be a new genetic factor for systemic lupus erythematosus (SLE). Linkage was shown by an overtransmission of the rs2004640-T allele from heterozygotic parents to patients with SLE (P<.001). The replicated association showed an allele frequency increasing from 0.51 in the general population to 0.61 in the patient population, with a risk conferred by the homozygotic T/T genotype greater than that of the heterozygotic genotype (P<.001). SLE and rheumatoid arthritis, which show some degree of familial aggregation, recently have been found to share one genetic factor, the PTPN22-1858T allele, which is also shared by several other autoimmune diseases. Estimated from the overtransmission observed in the authors’ trio sample with rheumatoid arthritis (61%), the estimated allele sharing in affected siblings would be 52 percent, which is virtually undetectable with the available sample. These results prompted the authors to hypothesize that the SLE IRF5 factor could also be associated with rheumatoid arthritis. The absence of linkage suggestion in genome scans at its location 7q32 was not sufficient, as shown by PTPN22, to exclude it as a rheumatoid arthritis gene. The authors conducted a study to test this hypothesis through linkage and association. One hundred French Caucasian trio families with rheumatoid arthritis were genotyped and analyzed with the transmission disequilibrium test, the frequency comparison of the transmitted and untransmitted alleles, and the genotype relative risk. Ninety-seven percent power was available to detect at least a trend in favor of a factor similar to that reported for SLE. The analysis showed the absence of linkage and association globally and in autoimmune rheumatoid arthritis subsets, with a weak nonsignificant trend against the IRF5 rs20046470-T allele. Given the robustness of a familial-based analysis, this slight negative trend provided strong evidence against even a weaker factor than that reported for SLE. The authors concluded that the results exclude the IRF5 rs2004640-T allele as a major genetic factor for rheumatoid arthritis in this French Caucasian population. Garnier S, Dieudé P, Michou L, et al. IRF5 rs2004640-T allele, the new genetic factor for systemic lupus erythematosus, is not associated with rheumatoid arthritis. Ann Rheum Dis. 2007;66:828–831. Reprints: S. Garnier, GenHotel-EA3886, Laboratoire Européen de Recherche pour la Polyarthrite Rhumatoide, 2 rue Gaston Crémieux, 91057 Evry-Genopole Cedex, France; garniersophie@wanadoo.fr
Evidence indicates that apolipoprotein B-100 (apo B) is better than total cholesterol or low-density lipoprotein cholesterol at predicting coronary risk. This would be expected because a single apo B molecule is found in each molecule of all β-lipoproteins, making it a better measure of the number of potentially atherogenic particles, especially small, dense, low-density lipoprotein. Serum non-high-density lipoprotein cholesterol (non-HDLC) is better correlated with apo B than is low-density lipoprotein cholesterol (LDLC). Non-HDLC has been proposed as a surrogate for apo B. On the basis of relative hazard ratios, an article cited by the author concluded that non-HDLC was as good a predictor of future cardiovascular events in women, if not better, as apo B. However, another study, on the basis of P values, found that apo B was superior in men and women. Based on relative odds and risk ratios, additional studies concluded that both were strong predictors of coronary heart disease but that apo B was the better marker in men. Nevertheless, it is difficult to assess the possible clinical usefulness of one test versus the other from these studies. Relative risk, hazard and odds ratios, or other correlations based on statistically significant P values do not translate into diagnostic sensitivity and specificity values that are required for defining clinically useful discrimination. The authors conducted a study in which the ability of apo B to discriminate between disease and nondisease was compared with non-HDLC and other lipoprotein lipids in 437 men who had undergone coronary angiography and were classified as having severe coronary artery disease (CAD), intermediate CAD, or no appreciable disease (normal). The ability of each test to discriminate was evaluated by three statistical techniques. First, stepwise discriminant analysis was used to determine how apo B compared with lipoprotein lipids across the entire group. Second, multivariate logistic regression was used to determine whether, when compared with data for subjects having no disease, data for subjects with only severe disease gave rise to the same conclusions as for the entire group. Logistic regression allows this study to be compared with other studies, most of which used logistic techniques for analysis. Third, these data were analyzed by receiver operating characteristic (ROC) curves that provide a means for definitive comparisons for clinical purposes before and after multivariate adjustment for nonlipoprotein risk factors. When analyzed by ROC curves, the difference between apo B and lipoprotein lipids proved to be less than would be anticipated from the odds ratios. Although, after adjustment, the difference was about 14 percent by odds ratios, ROC analysis showed a difference of only about one percent. This indicates that clinical studies should analyze the data using an absolute measure of risk, such as ROC curves, rather than just relative indexes. Such a small absolute difference may also explain discrepancies between studies. Levinson SS. Comparison of apolipoprotein B and non-high-density lipoprotein cholesterol for identifying coronary artery disease risk based on receiver operating curve analysis. Am J Clin Pathol. 2007;127:449–455 Reprints: Dr. Stanley S. Levinson, Laboratory Service, Dept. of Veterans Affairs Medical Center, 800 Zorn Ave., Louisville, KY 40206
Early graft function in renal transplant recipients has been demonstrated to be a predictor of long-term allograft survival. It may be affected by many immunological and nonimmunological factors, such as anti-leukocyte antibody titer, ABO compatibility, and ischemia-reperfusion injury. Furthermore, renal graft mass has been a determinant of outcome after kidney transplantation. To prevent hyperfiltration of the renal allograft, it is important to initially provide an adequate functioning nephron number to meet the metabolic demands of a recipient and to allow adequate nephron mass in case of rejection episodes, nephrotoxic drugs, or other potential injuries. During the preoperative evaluation of a potential kidney donor, it is necessary to estimate the renal function of the donated kidney compared with the metabolic needs of the recipient. The authors conducted a study in which they determined the correlation between the preoperative function of the donated kidney and the postoperative graft function of the recipient at one year post-transplantation. They measured the functional ratio of each kidney using technetium-99m diethylenetriaminepentaacetic acid. The serum creatinine (mg/dL) and estimated creatinine clearance (mL/ min./ 1.73 m2) using the Cockcroft-Gault formula were measured and calculated in 82 donors. The authors excluded recipients who had an episode of rejection and followed all recipients for more than six months post-transplantation. The average functional proportion of the donated kidney was 50.5±4.7 percent of the total creatinine clearance 83.4±18.3 of donors. The serum creatinine of recipients at one, three, six, and nine months post-transplantation was significantly (P<.05) correlated with the fractional creatinine clearance of the donated kidney; however, serum creatinine at one year was not correlated (P=.307). Furthermore, the creatinine clearance of the recipients at one, three, and six months post-transplantation was significantly (P<.05) correlated with the fractional creatinine clearance of the donated kidney; however, the creatinine clearance at nine months and one year was not correlated (P=.094 and .141, respectively). The authors concluded that the serum creatinine and creatinine clearance of recipients within six months after transplantation may depend on the functional mass of the donated kidney, which should be estimated prior to kidney donation and compared with the metabolic demands of the potential recipient. Lee BM, Yoon SN, Oh CK, et al. Fractional creatinine clearance of the donated kidney using Cockcroft-Gault formula as a predictor of graft function after living donor transplantation. Transplant Proc. 2006; 38: 1974– 1976. Reprints: Dr. Chang-Kwon Oh, Dept. of Surgery, Ajou University School of Medicine, 5 Wonchon-Dong, Yeongtong-Gu, Suwon, 443-721, Korea; ohck@ajou.ac.kr
Radiographic contrast media-associated nephrotoxicity (CAN) is responsible for approximately 11 percent of iatrogenic renal insufficiency and is the third most common cause of hospital-acquired renal failure. CAN is traditionally defined as an increase in serum creatinine of 0.5 mg/dL or 25 percent from baseline within 72 hours of exposure. Given that the majority of patients undergoing invasive cardiovascular procedures are outpatients or are likely to be discharged within 24 hours after the procedure, assessing changes in serum creatinine after 24 hours often is problematic. Therefore, a practical means of predicting an early postprocedural increase in creatinine would be of clinical benefit. In pharmacokinetic terms, systemic exposure to a drug (and, therefore, its safety and efficacy profile) is best quantified by the area under the drug’s blood concentration versus time curve. The derivation and validation of the ratio of the volume of contrast media to the creatinine clearance (V/CrCl) as a surrogate for the area under the curve recently has been reported in healthy volunteers. The authors hypothesized that this index would be useful in predicting an abnormal increase in post-percutaneous coronary intervention (PCI) creatinine and that it would show better predictive performance over considerations of volume or underlying renal function alone. They tested this hypothesis prospectively in an unselected population of patients undergoing PCI, calculating V/CrCl in 3,179 consecutive patients. An increase in serum creatinine of more than 0.5 mg/dL by 24 to 48 hours was considered abnormal. Receiver operating characteristic methods were used to identify the optimal sensitivity and specificity for the observed range of V/CrCl. The predictive value of V/CrCl for the risk of an early increase in creatinine was assessed using multivariable logistic regression. The authors found that the overall incidence of an abnormal, early increase in creatinine was 1.5 percent. The mean and median values of V/CrCl for patients with (mean, 5.2±4.4; median, 4.3; interquartile range, 2.7–6.0) and without (mean, 3.0±2.0; median, 2.5; interquartile range, 1.7–3.8) an early creatinine increase were each significantly (P<.001) different between groups. Furthermore, there was a significant association between V/CrCl and an early increase in creatinine (overall and trend, P<.001). The receiver operating characteristic curve analysis indicated that a V/CrCl ratio of 3.7 was a fair discriminator for the early creatinine increase (C-statistic, 0.69). After adjusting for other known predictors of post-PCI creatinine increase, V/CrCl of 3.7 or greater remained significantly associated with an early abnormal increase in serum creatinine (odds ratio, 3.84; 95% confidence interval, 2.0–7.3; P<.001). The authors concluded that a V/CrCl ratio of greater than 3.7 was a significant and independent predictor of an early abnormal increase in serum creatinine after PCI in this unselected patient population. Laskey WK, Jenkins C, Selzer F, et al. Volume-to-creatinine clearance ratio: a pharmacokinetically based risk factor for prediction of early creatinine increase after percutaneous coronary intervention. J Am Coll Cardiol. 2007;50:584–590. Reprints: Dr. Warren K. Laskey, Division of Cardiology, Dept. of Internal Medicine, University of New Mexico School of Medicine, MSC10-5550 1, University of New Mexico, Albuquerque, NM 87131; wlaskey@salud.unm.edu
Hypertensive disorders of pregnancy, particularly preeclampsia, are associated with significant morbidity and mortality, especially when preeclampsia occurs before 34 weeks’ gestation. The symptoms of the disorder are well characterized and generally present in the late second to third trimester. However, the underlying pathology is present at the early stages of pregnancy, and there is evidence that it is associated with a failure of trophoblastic invasion of the maternal spiral arteries. A variety of proteins and hormones have been studied as potential early markers for preeclampsia. The authors investigated whether measuring maternal serum placental protein-13 (PP-13) and pregnancy-associated plasma protein-A (PAPP-A) at 11+0 to 13+6 weeks’ gestation alone or in combination with second-trimester uterine artery pulsatility measured by Doppler velocimetry is useful in predicting those women who will develop preeclampsia. They conducted a nested case-control study of preeclampsia cases with controls matched for gestational age and storage time for the maternal serum. Samples were collected as part of a first-trimester prenatal chromosomal anomaly screening program and were routinely tested for PAPP-A. An examiner who was blinded to pregnancy outcome tested PP-13 using an enzyme-linked immunosorbent assay. All patients also underwent uterine artery Doppler flow velocimetry to measure the mean pulsatility index at 22 to 24 weeks’ gestation. There were 446 controls and 44 cases with early preeclampsia where delivery was induced prior to 35 weeks. There were also 44 cases with preeclampsia in which delivery was not induced before term. Median PP-13 levels for controls, all cases, and early preeclampsia cases were 176.9, 121.9, and 111.7 pg/mL, with multiples of the median (MoMs) of 1.00, 0.69, and 0.63, respectively (P<.001). PAPP-A MoMs were 1.00, 0.89 (P=.076), and 0.89 (P=.042), and mean pulsatility indexes were 1.0, 1.6 (P<.001), and 1.7 (P<.001) for controls, all cases, and early cases, respectively. Receiver operating characteristic (ROC) curve analysis for all cases or early cases versus controls yielded areas under the curve for PP-13, PAPP-A, and pulsatility index, respectively, of 0.68 (95% CI, 0.61–0.74; P<.001), 0.56 (95% CI, 0.49–0.63; P=.076), and 0.79 (95% CI, 0.72–0.87; P<.001) for all cases and 0.71 (95% CI, 0.63–0.79; P<.001), 0.59 (95% CI, 0.51–0.68; P=.076), and 0.86 (95% CI, 0.77–0.94; P<.001) for early cases. Combining PAPP-A with PP-13 and pulsatility index did not add significantly to the sensitivity. The authors concluded that first-trimester PP-13 levels may be useful in predicting preeclampsia and early preeclampsia, and the accuracy of the method increases when coupled with second-trimester Doppler pulsatility index measurement. First-trimester PAPP-A provides some prediction for preeclampsia when combined with pulsatility index but does not add to the prediction of early preeclampsia when PP-13 and pulsatility index are used together. Additional studies are necessary to establish the value of PP-13 in first-trimester screening for preeclampsia. Spencer K, Cowans NJ, Chefetz I, et al. First-trimester maternal serum PP-13, PAPP-A and second-trimester uterine artery Doppler pulsatility index as markers of pre-eclampsia. Ultrasound Obstet Gynecol. 2007;29:128–134. Reprints: K. Spencer, Dept. of Clinical Biochemistry, Harold Wood Hospital, Gubbins Lane, Romford, RM3 0BE, United Kingdom; kevinspencer1@aol.com Dr. Bissell is Professor and Director of Clinical Services and Vice Chair, Department of Pathology, Ohio State University Medical Center, Columbus. |
