College of American Pathologists
Printable Version

  Clinical Abstracts





cap today

August 2003

Coombs-negative ABO hemolytic disease of the newborn
The A and B antigenic sites are relatively weak on the newborn red blood cell membrane, so there is relatively little anti-A or anti-B antibody on the neonatal RBC in ABO hemolytic disease. Consequently, the cord blood direct antiglobulin (DAT) test, or Coombs’ test, is only weakly positive and may be negative unless a sensitive test is used. Many hospital laboratories use the gel test, which is more sensitive than techniques used previously for detecting immunoglobulin G coating of newborn red blood cells. Nonetheless, significant jaundice in DAT-negative neonates with ABO incompatibility is still frequently attributed to isoimmunization. In these cases, no other cause of jaundice is sought. Because unconjugated bilirubin levels can be elevated due to increased bilirubin production or reduced elimination, finding jaundice does not provide clear evidence of hemolysis. The authors undertook a study to determine the rate of hemolysis as determined by end-tidal carbon monoxide corrected for ambient air (ETCOc), which is a measure of bilirubin production, in a group of DAT-negative ABO-incompatible neonates, and compared these with DAT-negative ABO-compatible neonates. They found no significant difference between the mean 12-hour ETCOc levels in 60 DAT-negative ABO-incompatible neonates and 171 DAT-negative ABO-compatible neonates, despite a difference between the mean levels in 14 DAT-positive ABO-incompatible neonates and the DAT-negative groups. Four neonates in the DAT-negative ABO-incompatible group had elevated ETCOc levels, and two were diagnosed with a specific hematologic abnormality-glucose-6-phosphate dehydrogenase deficiency (G6PD) and elliptocytosis. The authors concluded that a cause other than isoimmunization should be sought in DAT-negative newborns with significant jaundice or increased bilirubin production, even if those newborns are ABO incompatible.

Herschel M, Karrison T, Wen M, et al. Isoimmunization is unlikely to be the cause of hemolysis in ABO-incompatible but direct antiglobulin test-negative neonates. Pediatrics. 2002;110:127-130.

Reprints: Marguerite Herschel, MC 1051, Dept. of Pediatrics, University of Chicago Children’s Hospital, 5841 S. Maryland Ave., Chicago, IL 60637; mhersche@

Use of C-reactive protein for detecting occult pediatric bacterial infections
It is difficult to diagnose occult bacteremia in children. Studies have established that 1.6 to eight percent of children who are three to 36 months old and have temperatures of 39°C or higher will have bacteremia. Discrimination on the basis of clinical findings has not been sufficiently accurate. The best laboratory predictors of bacteremia are the white blood cell count and absolute neutrophil count, which have sensitivities and specificities of 70 to 86 percent. C-reactive protein may be a valuable addition to this armamentarium because it helps distinguish systemic bacterial infections from viral infections in immunocompetent and immunodeficient children. The kinetics of CRP metabolism are rapid enough to allow it to function as a good marker of the rise and fall of the inflammatory component in children. The authors conducted a study to assess the utility of CRP in this context. They analyzed 256 children, ages three to 36 months, who were seen in an urban children’s hospital emergency department. The children had received complete blood cell counts and cultures as part of their evaluations and were prospectively enrolled from February 2000 through May 2001. The children were a median age of 15.3 months and had a median temperature at triage of 40°C. Twenty-nine cases of occult bacterial infection were identified, including 17 cases of pneumonia, nine cases of urinary tract infection, and three cases of bacteremia. White blood cell counts ranged from 3.6 to 39.1 ¥ 109/µL, and absolute neutrophil counts ranged from 0.56 to 28.16 ¥ 109/L. The median CRP level was 1.7 mg/dL, with a range of 0.2 to 43.3 mg/dL. Using logistic regression and receiver operator characteristic curve analysis, the authors determined the optimal cutoff point for CRP to be 4.4 mg/dL. This cutoff point achieved a sensitivity of 63 percent and specificity of 81 percent for detecting occult bacterial infection in this population. The authors found that an ANC cutoff point of 10.6 ¥ 109/L offered the best predictive model for detecting occult bacterial infection based on a single test. Adding CRP to the ANC did little to enhance diagnostic utility.

Isaacman DJ, Burke BL. Utility of the serum C-reactive protein for detection of occult bacterial infection in children. Arch Pediatr Adolesc Med. 2002;156:905-909.

Reprints: Dr. Daniel J. Isaacman, Division of Pediatric Emergency Medicine, Children’s Hospital of the King’s Daughters, 601 Children’s Lane, Norfolk, VA 23507;

Minimum numbers necessary for ISI calibration of point-of-care coagulation testing devices
The international sensitivity index calibration of prothrombin time systems requires performing PT on 60 patients stabilized with long-term warfarin treatment and 20 healthy subjects. A World Health Organization international reference preparation for thromboplastin is used with the manual PT technique to test the plasma from such samples in parallel with the local PT procedure on the same samples. A criterion of less than three percent coefficient of variation of the resultant calibration slope is recommended. All of this leads to great complexity given the large number of POC testing monitors that need to be compared. Therefore, a simplified method for calibration of these monitors is needed. One way to accomplish this would be to reduce the number of blood samples required for this calibration. The authors established the minimum numbers of fresh blood samples that should be used for such a calibration. They used a Monte Carlo Bootstrap technique to study two separate POC testing PT systems. The authors found in more than 50,000 calibrations that there was little effect on the mean ISI by reducing sample numbers to seven. There was, however, progressively less certainty regarding the reliability of the calibration under these circumstances. The precision of the calibrations and the INR deviation were not markedly affected by reducing the number of samples to half with the POC testing systems. But because ISI calibration with the two POC testing systems was less precise than conventional manual testing, the authors did not advise reducing the number of samples.

Poller L, Keown M, Chauhan N, et al. Minimum numbers of fresh whole blood and plasma samples from patients and healthy subjects for ISI calibration of CoaguCheck and RapidPointCoag monitors. Am J Clin Pathol. 2002;117:892-899.

Reprints: Dr. Leon Poller, ECAA Central Facility, School of Biological Sciences, University of Manchester, Manchester, England

Evaluation of allergic transfusion reactions
Allergic (urticarial) transfusion reactions are one of the most common complications of blood component transfusion. ATRs are usually benign, but severe reactions, such as anaphylactic/anaphylactoid manifestations or hypotension, or both, can occur. The authors of this study retrospectively evaluated 273 consecutive ATRs at one institution during a nine-year period. All reactions reported to the transfusion service were evaluated and reviewed by physicians expert in blood banking/transfusion medicine. During the study period, 1,613 adverse reactions to transfusion were reported. Of these, 273 (17 percent) were identified as allergic. Red blood cells were implicated in 123 (45.1 percent) ATRs; fresh frozen plasma in 66 (24.2 percent); platelets in 81 (29.7 percent); fresh frozen plasma and pooled platelets in two (0.7 percent); and pooled cryoprecipitate in one (0.3 percent). One ATR with hives was seen in a major ABO mismatch transfusion, and five ATRs were associated with autologous units of red blood cells. Severe reactions (anaphylactic or anaphylactoid) were observed in 21 (7.7 percent) patients. Seven (33.3 percent) of these 21 severe ATRs involved red blood cells, and 14 (66.7 percent) involved fresh frozen plasma or platelet transfusions. Nine (42.9 percent) of the 21 severe ATRs had associated hypotension. The overall incidence of anaphylactic or anaphylactoid reactions or severe allergic reactions was 1.3 percent. The clinical manifestations were quite variable, and various types of skin manifestations were observed. Skin rashes were often localized to various parts of the body and were sometimes generalized, but no consistent skin presentation was identified. Twenty-six (9.5 percent) patients did not have skin manifestations. The overall incidence of ATRs was estimated to be one in 4,124 blood component transfusions, and severe ATRs occurred in approximately one in 53,612 blood component transfusions. Considering only red blood cells, platelets (doses), and fresh frozen plasma transfusions, the overall incidence of ATRs was one in 2,338 transfusions, and severe reactions were seen in one in 30,281 transfusions. The authors concluded that ATRs are usually benign but can be associated with severe clinical manifestations and autologous transfusion and should be investigated thoroughly.

Domen RE, Hoeltge GA. Allergic transfusion reactions: an evaluation of 273 consecutive reactions. Arch Pathol Lab Med. 2003;127: 316-320.

Correspondence: Dr. Ronald E. Domen, Dept. of Pathology, H160, Penn State Milton S. Hershey Medical Center, P.O. Box 850, 500 University Drive, Hershey, PA 17033;

Perirectal cultures for VRE surveillance
The Centers for Disease Control and Prevention guideline for isolation recommends contact isolation for patients colonized or infected with "epidemiologically important microorganisms," including vancomycin-resistant Enterococcus. Among the CDC recommendations for VRE is active surveillance with perirectal cultures and isolation of patients found to be colonized. The authors conducted a study to determine whether the costs related to VRE bacteremia justify the costs of preventive measures. They analyzed results from 10,400 perirectal swabs taken on 54,052 patients admitted to two university hospitals. Inpatients considered to be at high risk for VRE at the first hospital underwent weekly perirectal surveillance cultures. The estimated cost of culture and resulting isolation during a two-year period were compared with the estimated excess cost from more frequent VRE bacteremias at the second hospital, which was similar in size and complexity but was not using surveillance cultures to control the spread of VRE throughout the hospital. Cultures and isolation cost approximately $253,099. At the first hospital, 193 (0.38 percent) patients were culture positive, and only one showed VRE bacteremia. At the second hospital, at which surveillance and isolation were not practiced, 29 VRE bacteremias were recorded. The estimated attributable cost of VRE bacteremias at the comparison hospital, which was $761,320, exceeded the cost of surveillance by threefold. The authors concluded that VRE surveillance, as recommended by the CDC, may be cost-effective.

Muto CA, Giannetta ET, Durbin LJ, et al. Cost-effectiveness of perirectal surveillance cultures for controlling vancomycin-resistant Enterococcus. Infect Control Hosp Epidemiol. 2002;23:429-435.

Reprints: Dr. Barry M. Farr, Box 473, University of Virginia Health System, Charlottesville, VA 22908

Outcomes in AIDS associated with adherence to M. avium complex prophylaxis, antiretroviral therapy
Newer, more effective antiretroviral regimens and prophylactic treatments for opportunistic infections have dramatically reduced rates of morbidity and mortality from HIV disease. Adhering closely to these treatment regimens is crucial to their success. Many studies have shown a strong association between adherence to antiretroviral therapy and successful suppression of HIV-1 RNA. Recent studies have also shown that HIV-1 RNA levels independently predict survival and the occurrence of opportunistic infections. In contrast, relatively little is known about the association between adherence to potent antiretroviral and opportunistic infection prophylactic regimens and outcomes such as AIDS progression or the development of opportunistic infections. The authors assessed the rate of adherence to antiretroviral and Mycobacterium avium complex (MAC) prophylactic regimens. They studied 643 patients who were enrolled in a trial of MAC prophylaxis. Forty-two percent of the patients reported that they were not adhering to the prophylactic regimen by week 56 of the study followup. Twenty-five percent of the patients reported that they were not adhering to the antiretroviral regimens. Not adhering to both regimens was associated with higher HIV-1 RNA levels and a significant increase in the risk of developing AIDS-defining complications or dying. The authors concluded that these results underscore the clinical significance of adhering to HIV therapy.

Cohn SE, Kammann E, Williams P, et al. Association of adherence to Mycobacterium avium complex prophylaxis and antiretroviral therapy with clinical outcomes in acquired immunodeficiency syndrome. Clin Infect Dis. 2002; 34: 1129-1136

Reprints: Dr. Susan E. Cohn, University of Rochester Medical Center, Infectious Disease Unit, 601 Elmwood Ave., Box 689, Rochester, NY 14642; susan_cohn@

Probability-based reference ranges for ratios of log-gaussian analytes
There may be complications in analyzing laboratory data when the lab result and its associated reference range must be converted or calculated from other test results and reference ranges from a single patient visit. The basic concept of the reference range involves upper and lower reference limits, which are estimated to enclose a specified percentage (generally 95 percent) of the values for a population from which the reference subjects have been drawn. These upper and lower reference limits are typically assumed to demarcate the estimated 2.5th and 97.5th percentiles of the underlying distribution of values, respectively. An example of the problem of the calculated result and reference range is provided in the translation between a white blood cell differential absolute count reference range and a WBC differential percentage reference range, using the total WBC reference range. A naïve method for determining this would calculate the upper and lower limits of the derived test from the upper and lower limits of the measured values using the same algebraic formula used for the derived measure. This method and any others that do not rely on probability-based transformations don’t maintain the distributional characteristics of the original reference ranges. The authors proposed a probability-based approach for the interconversion of reference ranges for ratios of 2 log-gaussian analytes. The method involves a simple algebraic formula for calculating the reference ranges of the derived measures while preserving the probability relationships. The nonparametric method and a parametric method that takes the log transformation estimate a reference range, and then exponentiates are provided as comparators. The authors showed, with example data, that the proposed method maintains the distributional characteristics of the transformed reference range measures, while the naïve method does not.

Trost DC, Hu M, Brailey AG, et al. Probability-based construction of reference ranges for ratios of log-gaussian analytes. Am J Clin Pathol. 2002;117:851-856.

Reprints: A.G. Brailey, Pfizer Global Research and Development, 50 Pequot Ave., New London, CT 06320

Carotenoid levels in kids
Carotenoids are compounds with vitamin A-like chemical structures and antioxidant properties that are found primarily in plants, especially fruits and vegetables. Five of these compounds-α-carotene, β-carotene, β-cryptoxanthin, lutein and zeaxanthin, and lycopene-are the principle chemical entities that have measurable concentrations in human blood. Epidemiologic studies suggest that carotenoid intake or circulating concentrations of these compounds are inversely correlated to all-cause mortality, cardiovascular disease, various cancers, insulin resistance, and other conditions. Consequently, knowledge of the distribution of concentrations of these compounds in the population at large is considered to be of some value. The authors examined the distributions and determinants of concentrations of the five aforementioned compounds in U.S. children and adolescents who participated in the third National Health and Nutrition Examination Survey (NHANES III). They conducted a cross-sectional study of 4,231 subjects who ranged in age from six to 16 years. The authors adjusted for age, gender, race or ethnicity, poverty-income ratio, body mass index status, and concentrations of high-density lipoprotein and non-HDL cholesterol, C-reactive protein, and cotinine. After these adjustments, only HDL cholesterol (P<0.001) and non-HDL cholesterol (P<0.001) concentrations were directly related to all of the carotenoid concentrations. Age (P<0.001) and body mass index status (P<0.001) were inversely related to all carotenoid concentrations, except those of lycopene. Young males had slightly higher carotenoid concentrations than did young females, but the differences were significant only for lycopene (P=0.029). Various ethnic differences were also found in the distributions of these compounds. C-reactive protein concentrations were inversely related to concentrations of b-carotene (P<0.001), lutein and zeaxanthin (P<0.001), and lycopene (P=0.023). Cotinine concentrations were inversely related to concentrations of α-carotene (P=0.002), β-carotene (P<0.001), and β-cryptoxanthin (P<0.001). The authors concluded that their findings may serve as valuable reference range information for these analytes in U.S. children and adolescents.

Ford ES, Gillespie C, Ballew C, et al. Serum carotenoid concentrations in U.S. children and adolescents. Am J Clin Nutr. 2002;76:818-827.

Reprints: E.S. Ford, Centers for Disease Control and Prevention, 1600 Clifton Rd., MS E-17, Atlanta, GA 30333;

A disease-susceptibility gene for coronary artery disease
Pseudoxanthoma elasticum is an inherited disorder involving dystrophic mineralization of elastic tissues of the skin, retina, and arterial walls. The frequency of PXE in the general population is not known because it is likely that individuals with a mild clinical phenotype escape diagnosis. The molecular basis of PXE appears to be a mutation in an ATP-binding cassette (ABC) transporter gene (ABCC6). Accelerated atherosclerosis, leading to myocardial infarction at a young age, appears to be a cardiovascular manifestation of PXE. PXE has, on several occasions, been an incidental finding in patients with premature cardiovascular disease. It has been impossible to assess whether being a carrier of a single ABCC6 gene mutation on one allele would confer additional risk for coronary artery disease. The authors conducted a case-control study involving 441 patients no older than 50 years of age who had definite CAD and 1,057 age- and gender-matched, population-based controls who were free of CAD to assess the relationship between the frequent R1141X mutation in the ABCC6 gene and the prevalence of premature CAD. The prevalence of the R1141X mutation was 4.2 times higher among patients than among controls. Therefore, subjects with the R1141X mutation had an odds ratio for a coronary event of 4.23, which represents a sharply increased risk of premature CAD.

Trip MD, Smulders YM, Wegman JJ, et al. Frequent mutation in the ABCC6 gene (R1141X) is associated with a strong increase in the prevalence of coronary artery disease. Circulation. 2002;106:773-775.

Reprints: Mieke D. Trip, Dept. of Cardiology, Academic Medical Centre, Meilbergdreef 9, 1105 AZ Amsterdam, Netherlands; m.d.trip@amc.uva.nln

Clinical pathology abstracts editors

Michael Bissell, MD, PhD, MPH, professor and director of clinical services and vice chair, Department of Pathology, Ohio State University Medical Center, Columbus.

Ronald Domen, MD, professor of pathology, medicine, and humanities, Penn State University College of Medicine, Hershey, Pa.