College of American Pathologists
Printable Version

  Clinical Abstracts





cap today

February 2006

Michael Bissell, MD, PhD, MPH
Ronald Domen, MD

bullet Breast carcinoma malignancy grading by Bloom-Richardson system vs. proliferation index

Questions remain regarding the reproducibility and efficacy of histologic malignancy grading relative to alternative proliferation index measurements for outcome prediction. Consequently, seven pathologists evaluated microsections of specimens from the Cooperative Breast Cancer Tissue Resource (CBCTR) for reproducibility of grade and classification. They assessed nuclear figure classification using photographs, and they assigned grade by Bloom-Richardson method, Nottingham modification. The authors evaluated proliferation index prospectively by deoxyribose nucleic acid precursor uptake with thymidine (autoradiographic) or bromodeoxyuridine (immunohistochemical) labeling index using fresh tissue from 631 node-negative breast cancer patients accessioned at St. Luke's Hospital, Chesterfield, Mo. A modified Nottingham-Bloom-Richardson grade was derived from histopathologic data. Median post-treatment observation was 6.4 years. Agreement on classification of nuclear figures (n=43) was less than good by kappa statistic (κ=0.38). Grade was moderately reproducible in four trials (n=10, 10, 19, 10) with CBCTR specimens (κ=0.50 to 0.59). Of components of Bloom-Richardson grade, agreement was least for nuclear pleomorphism (κ=0.37 to 0.50), best for tubularity (κ=0.57 to 0.83), and intermediate for mitotic count (κ=0.45 to 0.64). Bloom-Richardson grade was a univariate predictor of prognosis in node-negative St. Luke's patients and was improved when mitotic count was replaced by labeling index (low, mid, or high). When labeling index was added to a multivariate model containing tumor size and vessel invasion, grade was no longer a significant predictor of tumor-specific relapse-free or overall survival. Mitotic index predicted best when intervals were lowered to zero to two, three to 10, and more than 10 mitotic figures per 10 0.18-mm2 high-power fields. The authors concluded that Nottingham-Bloom-Richardson grades remain only modestly reproducible. Prognostically useful components of grade are mitotic index and tubularity. The Nottingham-Bloom-Richardson system can be improved by lowering cutoffs for mitotic index and by counting 20 to 30 rather than 10 high-power fields. Measurement of proliferation index by immunohistochemically detectable markers will probably give superior prognostic results in comparison to grade.

Meyer JS, Alvarez C, Milikowski C, et al. Breast carcinoma malignancy grading by Bloom-Richardson system vs proliferation index: reproducibility of grade and advantages of proliferation index. Mod Pathol. 2005;18:1067-1078.

Reprints: Dr. J.S. Meyer, Dept. of Pathology, St. Luke's Hospital, 232 S. Woods Mill Rd., Chesterfield, MO 63017;

bullet A clinicopathological study of mantle cell lymphoma

Mantle cell lymphoma is a distinct B cell lymphoma subtype with a particularly poor clinical outcome. The clinical relevance of the morphologic characteristics—that is, classic versus blastic cytology, diffuse versus nodular growth pattern, proliferation indices—of this tumor remains uncertain. The European MCL Network reviewed 304 cases of mantle cell lymphoma (MCL) to determine the prognostic significance of histopathologic characteristics. Cytomorphologic subtypes, growth pattern, and markers of proliferation were analyzed. Most cases fell into the cytologic subtypes of classical (87.5 percent), which is defined as monotonous small to medium-sized cells with indented nuclei, scant cytoplasm, and moderately dispersed chromatin; small cell (3.6 percent), which is defined as small cells with round or slightly indented nuclei and dense chromatin, mimicking CLL/SLL; pleomorphic (5.9 percent), which is defined as medium to large cells with nuclei of different sizes and shapes and moderate amounts of cytoplasm, mimicking DLBCL; and blastic (2.6 percent), which is defined as monomorphic small to medium-sized cells with dispersed chromatin, scant cytoplasm, and frequent apoptotic bodies, mimicking lymphoblastic lymphoma. The network also identified two subgroups with mixtures of cells (classical + pleomorphic; 1.6 percent) or transitions between cell types (classical/pleomorphic; 1.6 percent). There was no statistically significant difference in overall survival among the histologic groups, but the number of pleomorphic and blastic cases was small. A diffuse growth pattern was seen in 80.5 percent of cases, while 19.5 percent had a nodular or mantle zone pattern, which was associated with a slightly better prognosis (median overall survival, 43 months versus 29 months for diffuse pattern; P=0.0074). A high proliferation rate by mitotic or Ki-67 indices was associated with shorter overall survival. Cut-off levels were defined that discriminated three subgroups based on proliferation rates—less than 10 percent, 10 percent to 40 percent, and greater than 40 percent—that showed significantly different clinical outcomes. In a univariate analysis of all prognostic factors with respect to overall survival, Ki-67 and mitotic indices, growth pattern, age, bone marrow involvement, performance status, LDH level, stage, B symptoms, and extranodal involvement resulted in significant differences in overall survival. The authors concluded that, based on this large clinicopathologic study, multivariate analysis confirmed the central prognostic role of cell proliferation and its superiority to other histologic or clinical criteria.

Tiemann M, Schrader C, Klapper W, et al. Histopathology, cell proliferation indices and clinical outcome in 304 patients with mantle cell lymphoma (MCL): a clinicopathological study from the European MCL Network. Br J Haematol. 2005;131:29-38.

Correspondence: Dr. C. Schrader, Dept. of Haematology, University Hospitals of Schleswig-Holstein, Campus Kiel, Chemnitzstr. 33, 24116, Kiel, Germany

bullet Perineural involvement by benign prostatic glands on needle biopsy

Perineural invasion by prostatic glands in needle-biopsy specimens is rarely used as a diagnostic marker of prostatic adenocarcinoma; however, its presence in a case with carcinoma may influence whether or not nerve-sparing surgery is performed. The authors conducted a study of benign perineural involvement in which they described 27 needle-biopsy cases of benign prostatic glands with perineural/intraneural involvement. The benign glands indented nerves in 14 cases, tracked along nerves in eight cases, and wrapped around them in seven cases. The benign glands were intraneural in four cases and adjacent to cross-sections of nerves in two cases. In one of the seven cases with a pattern of wrapping around nerves, the benign glands involved approximately 95 percent of the circumference. Immuno-staining (by high-molecular-weight cytokeratin and p63) for basal cells was performed in eight cases and was positive in six. In the two cases that were negative for both markers, the glands had features of partial atrophy. The presence of benign features, such as papillary infolding, ample pale to clear cytoplasm, small bland nuclei, absence of crystalloids, and recognition of basal cells on hematoxylin-and-eosin staining, are clues to the correct interpretation. Since partial atrophy may be negative for basal cell markers, as seen in the two cases in this study, it may pose a particular diagnostic problem. However, as the authors noted, comparison of such acini to adjacent and distant benign glands in the same biopsy, along with careful attention to morphology, are helpful in ruling out carcinoma. The authors concluded that it is important to recognize this phenomenon and its patterns to avoid diagnostic pitfalls.

Ali TZ, Epstein JI. Perineural involvement by prostatic glands on needle biopsy. Am J Surg Pathol. 2005;29:1159-1163.

Reprints: Dr. Jonathan I. Epstein, Johns Hopkins Hospital, Dept. of Pathology, 401 N. Broadway St., Room 2242, Weinburg Bldg., Baltimore, MD 21231;

bullet Lymphocyte-neutrophil index in FNA biopsy samples as a prognostic factor for advanced NSCLC

The degree of infiltration by lymphocytes within resected nonsmall cell lung carcinoma correlates significantly with survival. However, it is not known whether this holds true for unresectable advanced stage tumors. Therefore, the authors determined the amount of lymphocytes within aspiration smears from 147 advanced stage (stages IIIB and IV) nonsmall cell lung carcinoma (NSCLC) cases and correlated it with survival rate. Two groups were categorized as lymphocyte dominant and nondominant on the basis of the lymphocyte-neutrophil (L-N) index. This index was calculated as the difference between the lymphocyte-neutrophil ratio in the smear and in peripheral blood obtained on the same day as the procedure. The ratio is the number of lymphocytes to the sum of the number of lymphocytes and neutrophils. The lymphocyte-dominant group had an L-N index greater than or equal to the cut-off value as determined from a control group of 41 smears considered to be contaminated by abundant blood. The survival rate was significantly higher in the lymphocyte-dominant group, and the L-N index was an independent prognostic factor. The L-N index-based grouping also correlated significantly with the histologic grade of tumor-infiltrating lymphocytes in 164 cases of resected stage IA-IIIA NSCLC. The authors concluded that determining the L-N index in aspiration smears of unresectable NSCLC appears to be a reproducible method and an independent prognostic factor.

Nakahara Y, Mochiduki Y, Miyamoto Y, et al. Prognostic significance of the lymphocyte-to-neutrophil ratio in percutaneous fine-needle aspiration biopsy specimens of advanced nonsmall cell lung carcinoma. Cancer. 2005;104:1271-1280.

Reprints: Dr. Yasuharu Nakahara, Dept. of Respiratory Medicine, National Hospital Organization Himeji Medical Center, 68 Honmachi, Himeji-city 670-8520, Japan;

bullet Value of CD8, CD4, CD68, HLA-DR cells in a prognosis of cutaneous malignant melanoma with vertical growth

To establish the prognostic value of immune system cells that infiltrate melanoma, the authors evaluated the distribution and density of T lymphocyte subsets, macrophages, and dendritic cells in samples of primary cutaneous melanoma from 47 patients with stages I and II melanoma according to the American Joint Committee on Cancer staging system. Immunohistochemical demonstrations of CD8 and CD4 lymphocytes, CD68 macrophages, human leukocyte antigen-D-related (HLA-DR) cells, S-100 protein, and melanoma-associated antigens Melan-A and HMB-45 were performed. The results were derived from independent histopathologic reviews by two pathologists. The low-density, moderate-density, and high-density groups of cells that infiltrate the base of the tumor during the vertical growth phase were compared with the overall survival rate using the Kaplan-Meier method and the log-rank test. Clinical variables (gender, age, tumor location, Clark level, vascular/lymphatic invasion, and thickness) also were analyzed. The CD8 lymphocytes exhibited independent, statistically positive significance in survival (log-rank test, 8.49; P=0.01) between patients in different lymphocyte density groups. The five-year survival rate differed among patients in the high-density group (78.8 percent), moderate-density group (44.4 percent), and low-density group (25 percent). The CD4 lymphocytes, which were less numerous than CD8 cells, had similar distribution. A correlation was also noted between HLA-DR cells and overall survival (log-rank test, 5.29; P=0.02). CD68 cell density did not correlate with survival. The authors concluded that the presence and number of infiltrating CD8 lymphocytes, as well as the overall occurrence of HLA-DR cells, may be considered independent, favorable prognostic factors in melanoma. The results may be important for identifying other prognostic factors with which to evaluate disease progression and develop immune therapies for patients with melanoma.

Piras F, Colombari R, Minerba L, et al. The predictive value of CD8, CD4, CD68, and human leukocyte antigen-D-related cells in the prognosis of cutaneous malignant melanoma with vertical growth. Cancer. 2005;104:1246-1254.

Reprints: Dr. Paola Sirigu, Dept. of Cytomorphology, University of Cagliari, Cittadella Universitaria, 09042 Monserrato (CA), Italy;

bullet Variability in cytologic-histologic correlation practices and implications for patient safety

The Clinical Laboratory Improvement Amendments of 1988 require that laboratories perform cytologic-histologic correlation, although the optimal methods for and value of performing correlation have not been determined. To assess the similarities and differences in how laboratories perform cytologic-histologic correlation, the authors sent a letter to 162 laboratories in the United States requesting copies of the materials they used in the cytologic-histologic correlations process. The returned materials were classified into the categories of forms, logs, and tally sheets. A checklist (derived from the College of American Pathologists’ Laboratory Accreditation Program cytopathology checklist) was developed to classify the “minimum expected” (15) and “additional” data points that laboratories collected when they performed a correlation. The authors then measured the percentage of laboratories that recorded minimum expected and additional data points and the frequency with which specific minimum expected data points were recorded. The response frequency was 32.1 percent, and the authors obtained 84 cytologic-histologic correlation materials. The only minimum expected variables recorded on forms or logs by more than 50 percent of laboratories were cytology case number, sign-out cytology diagnosis, surgical pathology case number, and sign-out surgical pathology diagnosis. Nine (17.3 percent) laboratories did not record data on forms, logs, or tally sheets. The mean number of minimum expected and additional variables recorded on forms was 6.5 and 8.7, respectively. The authors concluded that laboratories record data from the cytologic-histologic correlation process in a number of ways, indicating that the data-collection process lacks standardization.

Vrbin CM, Grzybicki DM, Zaleski S, et al. Variability in cytologic-histologic correlation practices and implications for patient safety. Arch Pathol Lab Med. 2005;129:893-898.

Reprints: Dr. Stephen S. Raab, Dept. of Pathology, University of Pittsburgh, 5150 Centre Ave., Ste. 301, Pittsburgh, PA 15232;

bullet Pancreatic ductal adenocarcinomas with cystic features

Cystic tumors of the pancreas are uncommon but important because of their diverse pathology and biology. Their wide spectrum also includes cystic variants of otherwise solid tumors, such as cystic endocrine tumors, cystic acinar cell carcinomas, and ductal adenocarcinomas with cystic changes. The authors of this study screened pancreatic ductal carcinomas and their variants for macrocystic changes and determined the nature of the cysts (neoplastic versus non-neoplastic). Of 483 tumors, 38 (eight percent) had cystic features. The largest group consisted of 24 pancreatic ductal adenocarcinomas showing a large-gland pattern with small cysts that had diameters varying between 0.5 cm and 1.8 cm. The epithelial lining of these cysts was generally positive for CEA (83 percent) and/or MUC1 (71 percent) and MUC5AC (74 percent). P53 was positive in 57 percent of the cases. The second group of cystic tumors (8/483) showed degenerative cystic cavities with diameters between 1 cm and 6 cm. This group consisted of poorly differentiated pancreatic ductal adenocarcinomas, undifferentiated carcinomas with or without osteoclast-like giant cells, and one adenosquamous carcinoma. The third group of cystic tumors had four pancreatic ductal adenocarcinomas containing tumor-related retention cysts. Their epithelial cells were positive for MUC5AC but negative for CEA MUC1 and p53. The fourth group consisted of two pancreatic ductal adenocarcinomas showing closely attached pseudocysts caused by tumor-associated pancreatitis. The results of the study indicate that a considerable number of pancreatic ductal adenocarcinomas and their variants display cystic features and must, therefore, be considered in the differential diagnosis of cystic neoplasms of the pancreas. Moreover, not all of the cystic structures observed were neoplastic in nature. They may also represent non-neoplastic changes, such as retention cysts and inflammatory pseudocysts.

Kosmahl M, Pauser U, Anlauf M, et al. Pancreatic ductal adenocarcinomas with cystic features: neither rare nor uniform. Mod Pathol. 2005;18:1157-1164.

Reprints: Dr. M. Kosmahl, Dept. of Pathology, University of Kiel, Michaelisstr. 11, Kiel 24105, Germany;

bullet Coagulase-negative staph in blood cultures

Coagulase-negative staphylococci are the pathogens most commonly reported to cause central vascular line-associated bloodstream infection, accounting for 38 percent of such infections reported to the Centers for Disease Control and Prevention's National Nosocomial Infections Surveillance system from 1995 to 2002 (CDC, unpublished data). However, coagulase-negative staphylococci (CoNS) are also common skin commensals that frequently contaminate blood cultures. The misinterpretation of CoNS contaminants as indicative of true bloodstream infections has implications for patient care and hospital quality assurance. Regarding patient care, unnecessary use of antimicrobials, especially vancomycin, raises health care costs, selects for antimicrobial-resistant organisms, and exposes the patient to possible adverse drug effects. Regarding quality assurance, monitoring the rate of catheter-associated bloodstream infections is widely recommended and practiced. However, the commonly used definition of bloodstream infection may have limited capacity to exclude contaminants, resulting in inaccurate surveillance data and exaggerating the role of CoNS in bloodstream infection. The CDC's campaign to prevent antimicrobial resistance includes a recommendation to treat infection, not contamination, and suggests that cultures positive for CoNS be interpreted on the basis of the number of positive results among the number of cultures performed. To help interpret cultures, the CDC constructed a mathematical model of blood cultures positive for CoNS in patients with a central vascular line. To simulate the likely variability among institutions and patient populations, results over a range of possible input values were shown. Predictive values for blood cultures positive for CoNS in patients with a central vascular line in place were calculated using the following rates: true bacteremia, three percent; blood culture contamination, two percent; detection of bacteremia, 80 percent; and catheter colonization, two percent (for blood samples obtained through a central vascular line). Positive predictive values were 55 percent for one positive culture result of one culture performed, 20 percent for one positive result of two performed, and five percent for one positive result of three performed. For two positive culture results of two cultures performed, the positive predictive value was 98 percent if both samples were obtained through the vein, 96 percent if one sample was obtained through a catheter and the other was obtained by vein, and 50 percent if both samples were obtained through a catheter. The author concluded that using this model with institution-specific values for input parameters would assist in clinical decisionmaking as well as hospital quality assurance.

Tokars JI. Predictive value of blood cultures positive for coagulase-negative staphylococci: implications for patient care and health care quality assurance. Clin Infect Dis. 2004;39:333-341.

Reprints: Dr. Jerome I. Tokars, Centers for Disease Control and Prevention, 1600 Clifton Rd., Mailstop E-55, Atlanta, GA 30333;

bullet Urine myoglobin determinations

Tests used to evaluate rhabdomyolysis include serum myoglobin, serum creatine kinase, urine myoglobin, and urine heme pigments. Early, aggressive management of rhabdomyolysis has been shown to limit development of acute renal failure, so prompt and accurate diagnosis is essential. The test most commonly used in the diagnosis of rhabdomyolysis is serum creatine kinase (CK). Myoglobin, a 16.9 kDa oxygen-binding hemoprotein, is found in all striated muscles and appears in the circulation within a few hours after cardiac or skeletal muscle damage. Testing for serum myoglobin (sMb) is thought to be less sensitive but more specific than serum CK for muscle breakdown because it may be rapidly eliminated by renal and extra-renal mechanisms. Identification of myoglobin in the urine by centrifugation through a microconcentrator membrane has been described to qualitatively assess the presence of urine myoglobin. Using this technique, urine myoglobin (16.9 kDa) and hemoglobin (64 kDa) can be separated by centrifugation through a microconcentrator membrane with a 30 kDa cutoff. This separation allows for individual analysis of the hemoglobin-free urine filtrate for the presence of heme pigment—that is, myoglobin—using a peroxidase-based dipstick. In the authors' practice, when evaluating patients with clinically diagnosed rhabdomyolysis, this qualitative assay frequently provides negative results, which led the authors to question its accuracy and utility as a screening test. The authors conducted a retrospective analysis of the relationship between CK, sMb, the urine qualitative assay for myoglobin, and the semi-quantitative assay for urine heme pigments (uH) in patients evaluated for rhabdomyolysis. There were 673 patients with CK and urine myoglobin (uMb) recorded on the same day. The uMb assay has a sensitivity of only 26.4 percent (95 percent confidence interval [CI], 23.1 to 29.7 percent) and specificity of 96.8 percent (95 percent CI, 95.5 to 98.1 percent) for detecting severe rhabdomyolysis, defined as a CK greater than 10,000 U/L. Serum myoglobin and CK measured simultaneously in 83 patients were highly correlated (r2=0.72 for log-transformed values), suggesting that the negative uMb test was not a result of the absence of sMb. In 241 patients who had CK, uMb, and uH measured on the same day, the presence of moderate or large uH in the absence of hematuria, indicating the presence of myoglobinuria, had a sensitivity of 81 percent (95 percent CI, 76 to 86 percent) for the detection of CK greater than 10,000 U/L versus a sensitivity of 22 percent (95 percent CI, 17 to 27 percent) for the uMb assay. The authors concluded that the microconcentrator-based uMb assay has a poor and clinically inadequate sensitivity for detecting and diagnosing rhabdomyolysis.

Grover DS, Atta MG, Eustace JA, et al. Lack of clinical utility of urine myoglobin detection by microconcentrator ultrafiltration in the diagnosis of rhabdomyolysis. Nephrol Dial Transplant. 2004;19:2634-2638.

Reprints: Dr. Derek M. Fine, Division of Nephrology, Johns Hopkins University School of Medicine, 1830 E. Monument St., Suite 416, Baltimore, MD 21205;

bullet Relationship between salt intake and blood pressure

A recent report from the World Health Organization showed that elevated blood pressure is responsible for 62 percent of stroke and 49 percent of coronary heart disease worldwide. Much evidence from epidemiological, migration, intervention, genetic, and animal studies suggests that salt intake plays an important role in regulating blood pressure. Furthermore, there is increasing evidence that a high salt intake has direct harmful effects on the cardiovascular system—for example, it increases the mass of left ventricular wall, stiffens conduit arteries, and thickens and narrows resistance arteries independent of and additive to the effect of salt on blood pressure. However, the mechanisms whereby salt raises blood pressure and induces direct cardiovascular organ damage are not clear. Much evidence suggests that in those who develop high blood pressure, there is an underlying defect in the ability of the kidney to excrete salt, and that the greater compensatory response required to restore sodium balance is responsible for the increase in blood pressure. However, the potential role of small changes in plasma sodium has rarely been considered. To study this further, the authors reanalyzed three types of studies that they conducted on changing salt intake. In the first, an acute and large reduction in salt intake from 350 mmol/d to 10 mmol/d to 20 mmol/d for five days in hypertensive and normotensive subjects was associated with a fall in plasma sodium of approximately 3 mmol/L (P<0.001). In the second, progressive increases in salt intake from 10 mmol/d to 250 mmol/d by a daily amount of 50 mmol in normotensive subjects caused increases in plasma sodium (P<0.001). In the third, a longer-term modest reduction in salt intake in hypertensive subjects was studied in double-blind, randomized crossover studies—one month of usual salt intake (≈170 mmol/d) compared with reduced salt intake (≈100 mmol/d). There was a decrease in plasma sodium of 0.4±0.2 mmol/L (P<0.05), which was weakly but significantly correlated with the fall in systolic blood pressure (r=0.18, P<0.05). These studies demonstrate that an increase or a decrease in salt intake causes changes in plasma sodium. The authors concluded that small changes in plasma sodium alter extracellular volume, which may influence blood pressure. Changes in plasma sodium may also directly affect blood pressure.

He FJ, Markandu ND, Sagnella GA, et al. Plasma sodium—ignored and underestimated. Hypertension. 2005;45:98-102.

Reprints: G.A. MacGregor, Blood Pressure Unit, St. George's Hospital Medical School, Cranmer Terrace, London, SW17 ORE, United Kingdom;

bullet New specimen types for cortisol measurements

Endocrinologists managing hypoadrenal patients face the challenge of determining the appropriate dose for physiological glucocorticoid replacement. Factors that may account for the wide range in dosages include differences in hydrocortisone adsorption, variations in tissue sensitivity to cortisol, and other medications that interfere with cortisol metabolism. Various laboratory methods have been used to assess the adequacy of glucocorticoid replacement therapy, but none is satisfactory. The cortisol “day curve,” which involves measuring serum cortisol levels at various time points, is useful for monitoring replacement therapy, but it requires multiple blood tests at a pathology collection center that are time-consuming, invasive, and inconvenient for the patient. As a result, it has not been adopted routinely by clinicians. The authors conducted a study to explore alternative cortisol sampling methods that are less invasive and permit ambulatory testing by patients. Because cortisol can be measured in saliva as well as in capillary blood in the form of blood spots, the authors sought to determine whether these sampling methods could be used as reliable alternatives to venipuncture. They first examined the correlations between serum cortisol, saliva, and blood spot cortisol in hospital inpatients who were not receiving glucocorticoid therapy. They then studied 18 hypoadrenal patients on hydrocortisone therapy and measured their serum, saliva, and blood spot cortisol concurrently at seven different time points in one day. For hospital inpatients, a significant correlation was noted between saliva and serum cortisol (r=0.7121, P<0.0001), but there was a stronger correlation between blood spot and serum cortisol (r=0.9494, P<0.0001). The correlations were weaker in hypoadrenal patients on hydrocortisone (saliva and serum cortisol, r=0.6262, P<0.001; blood spot and serum cortisol, r=0.7871, P<0.001). When the authors evaluated each measurement with respect to an arbitrary target range, there was a greater degree of agreement between serum and capillary blood spot cortisol (85 percent agreement) than between serum and saliva cortisol (65 percent agreement; P<0.001). The authors concluded that blood spot samplings provide a simple and convenient way for ambulant hypoadrenal patients on hydrocortisone replacement therapy to assess cortisol levels at multiple times in one day. This may be useful in determining the optimal glucocorticoid dose for hypoadrenal patients.

Wong V, Yan T, Donald A, et al. Saliva and bloodspot cortisol: novel sampling methods to assess hydrocortisone replacement therapy in hypoadrenal patients. Clin Endocrinol. 2004;61:131-137.

Reprints: Dr. Vincent Wong, Dept. of Diabetes and Endocrinology, Westmead Hospital, P.O. Box 533, Wentworthville, NSW 2145, Australia;

Dr. Bissell is Professor and Director of Clinical Services and Vice Chair, Department of Pathology, Ohio State University Medical Center, Columbus. Dr. Domen is Professor of Pathology, Medicine, and Humanities, Penn State University College of Medicine, Hershey, Pennsylvania.