Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Sampling Errors”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 433 records · Page 24Linked to original sources
PubMed2004

Completeness of ascertainment by cancer registries: putting bounds on the number of missing cases.

BACKGROUND: When comparing cancer incidence or mortality rates between different regions, it is important to know how complete the registration data are on which these figures are based. A number of ways of estimating completeness have been proposed, but it is often difficult to say how precise these estimates are. We describe a computer program developed to produce measures of precision for estimates of completeness obtained by one such method, the flow method. METHODS: The program works by resampling the required data sets, and repeatedly calculating completeness estimates until convergence of the standard errors occurs. It was tested on colorectal tumours from a single health district, and empirical confidence limits for 1 and 5 year completeness were compared with those obtained by applying various normalizing transformations and a beta distribution. The method was then applied to tumours of the head and neck, breast and lung and the results compared with those from a capture-mark-recapture exercise carried out 4 years previously. RESULTS: The sampling distribution was close to normal for 1 year completeness, but much less so for 5 year completeness, as assessed by quantile plots. Approximation by a beta distribution was better than by normalizing transformation. Although there were differences between the results produced by the flow method and capture-recapture, the flow method is more reproducible and easier to apply. CONCLUSION: It is now possible to estimate confidence limits for the results of the flow method, and thus determine whether comparative results between registries are likely to be affected by sampling error.

Bias↗
PubMed1985

Determination of organic acids in seven wheat varieties by capillary gas chromatography.

Fresh wheat tops were extracted with acidic 90% ethanol, and the ethanol was evaporated and a portion of the aqueous residue loaded onto DEAE-Sephadex. Organic acids were eluted with pyridinium formate and then lyophilized and the dried residue was derivatized with 1% trimethylchlorosilane in bis(trimethylsilyl)trifluoroacetamide. The acids were then quantitatively determined using capillary gas chromatography and identified using capillary gas chromatography-mass spectrometry. The acidic ethanol extraction of fresh plant tissue was quantitative for all acids except citric while losses in the remaining procedures were controlled by using an internal standard. The ion exchange chromatography made the greatest contribution to experimental error, imposing a minimum loading requirement of 0.1 mumol of each acid for adequate precision. Organic acid profiles were determined for seven wheat cultivars (Triticum aestivum cv Carazinho, Teal, Lance, Warigal, Isis, Maringa, and BH1146) grown on gravel in solution culture for 30 days. Profiles were simple, consisting of only malic, aconitic, and citric acids, with levels of each acid for all varieties falling within the range 2-5 mumol/g fresh tissue. Storage of samples led to a large increase in sampling error and increased the amount of extractable citric acid.

Calibration↗
PubMed1989

On the number of Purkinje cells in the human cerebellum: unbiased estimates obtained by using the "fractionator".

Stereological estimates of the numbers of Purkinje cell nucleoli in human cerebellar cortex have been obtained from systematic random samples of tissue by using the fractionator. The estimates are unbiased by fixation, section thickness, or sampling errors and are independent of any assumptions about cell shape, size, or spatial orientation. Twelve brains from aged subjects of both sexes were examined. The average complement of nucleoli in four female brains (age range 71-93 years) amounted to 14.8 millions (with an observed coefficient of variation between subjects of 29%). For three male brains (76-91 years), the corresponding estimates were 15.7 millions (10%). No significant sex differences were found for these small samples. Five brains of unknown sex and age yielded values of 15.8 millions (18%). For the twelve brains examined, the total number of Purkinje cell nucleoli per cerebellum was found to be 15.4 millions (19%). Estimated numbers showed a significant positive correlation with cerebellar weights. The number of nucleoli in an individual cerebellum was obtained with high precision in as short a time as 4 hours.

Aged↗
PubMed1990

Efficiency of marker-assisted selection in the improvement of quantitative traits.

Molecular genetics can be integrated with traditional methods of artificial selection on phenotypes by applying marker-assisted selection (MAS). We derive selection indices that maximize the rate of improvement in quantitative characters under different schemes of MAS combining information on molecular genetic polymorphisms (marker loci) with data on phenotypic variation among individuals (and their relatives). We also analyze statistical limitations on the efficiency of MAS, including the detectability of associations between marker loci and quantitative trait loci, and sampling errors in estimating the weighting coefficients in the selection index. The efficiency of artificial selection can be increased substantially using MAS following hybridization of selected lines. This requires initially scoring genotypes at a few hundred molecular marker loci, as well as phenotypic traits, on a few hundred to a few thousand individuals; the number of marker loci scored can be greatly reduced in later generations. The increase in selection efficiency from the use of marker loci, and the sample sizes necessary to achieve them, depend on the genetic parameters and the selection scheme.

Animals↗
PubMed1994

Genetic versus histological grading in stereotactic biopsies.

With diagnostic stereotactic biopsies in astrocytoma sometimes false results will be encountered. A false positive result is defined as a histological diagnosis which is afterwards proved to be untrue. Although this happens very seldom (1-2%) in characterizing the tumor type (e.g. astrocytoma versus lymphoma), there is a probability of undergrading in malignant astrocytomas due to regional heterogeneity in malignancy grade. The issue of 'sampling error' has been addressed by many authors. It was shown that undergrading in astrocytomas is a likely event in fibrillar astrocytomas that often show geographically distinct areas of well and poorly differentiated elements. In a recent study on histological grading versus genetic characterization of heterogeneous astrocytomas we published that low-grade areas within high-grade malignant astrocytomas have already the genetic features of high-grade malignancy. Therefore, particularly in stereotactic biopsies which contain relatively small samples of tumor tissue, this fact is of paramount importance. Undergrading may lead to withholding radiotherapy from the patient and to falsification of long-term survival results in series of so-called low-grade astrocytomas. We like to stress the importance of taking biopsies from different parts of the tumor, especially from parts with different density enhancement to contrast with CT scanning, and will discuss the technical advancements made in the genetic grading of astrocytomas. Particularly loss of heterozygosity of chromosome 10 and eventually amplification of the EGFR indicate a high grade of malignancy.

Adult↗
PubMed2003

Histopathologic analysis of atypical lesions in image-guided core breast biopsies.

Appropriate follow-up of patients with needle core breast biopsies (NCBB) showing atypical hyperplasia remains unclear because previous studies show that subsequent open biopsies in variable proportions of these patients reveal ductal carcinoma in situ (DCIS) or even invasive carcinoma, indicating significant sampling artifact. NCBB with diagnoses of atypia were morphologically classified into groups as follows: I, ALH (n = 24); II, ADH with minimal cytologic atypism (n = 90); III, atypia, other (9 columnar, 2 apocrine, 11 atypical papillary); IV, severe ADH/borderline DCIS (n = 31). Mammographic and histologic features, including the number of foci of atypia in the NCBB and the calcification span, were then correlated with presence of DCIS or invasive tumor in subsequent open excisions. Open excisional biopsies showed more severe lesions in 12% of Group I-III cases (8% in Group I, 9% in Group II, and 27% in Group III), of which 15 were DCIS and one was an invasive tubular carcinoma (0.3 cm). Of the DCIS, 60% (n = 9) were < or =5 mm, and 13 of 15 (87%) were low grade. The NCBB cavity was immediately adjacent to the more severe lesions in 88% (n = 14) of cases, in keeping with sampling error. The subset showing severe ADH with borderline nuclear features in contrast was associated with a high likelihood (63%) of DCIS in follow-up excisions. NCBB with atypical papillary features also showed a high frequency of DCIS (4/11, 36%) in subsequent open excisions. Other factors associated with more severe lesions on open biopsy included the number of atypical foci in the NCBB (>4, P <.05) and the mammographic calcification span (>2.0 cm, P <.0001). Atypical lesions diagnosed in NCBB samples are radiographically and morphologically heterogeneous, accounting for the variable frequency of DCIS or invasive neoplasm identified in subsequent open excisions, which are usually focal, low grade, and a consequence of sampling artifact (i.e., adjacent to the NCBB cavity). DCIS is more likely if microcalcifications are mammographically extensive or if atypia is multifocal or is associated with borderline cytologic features.

Biopsy, Needle↗
PubMed2004

A nearest-neighboring-end algorithm for genetic mapping.

MOTIVATION: High-throughput methods are beginning to make possible the genotyping of thousands of loci in thousands of individuals, which could be useful for tightly associating phenotypes to candidate loci. Current mapping algorithms cannot handle so many data without building hierarchies of framework maps. RESULTS: A version of Kruskal's minimum spanning tree algorithm can solve any genetic mapping problem that can be stated as marker deletion from a set of linkage groups. These include backcross, recombinant inbred, haploid and double-cross recombinational populations, in addition to conventional deletion and radiation hybrid populations. The algorithm progressively joins linkage groups at increasing recombination fractions between terminal markers, and attempts to recognize and correct erroneous joins at peaks in recombination fraction. The algorithm is O (mn3) for m individuals and n markers, but the mean run time scales close to mn2. It is amenable to parallel processing and has recovered true map order in simulations of large backcross, recombinant inbred and deletion populations with up to 37,005 markers. Simulations were used to investigate map accuracy in response to population size, allelic dominance, segregation distortion, missing data and random typing errors. It produced accurate maps when marker distribution was sufficiently uniform, although segregation distortion could induce translocated marker orders. The algorithm was also used to map 1003 loci in the F7 ITMI population of bread wheat, Triticum aestivum L. emend Thell., where it shortened an existing standard map by 16%, but it failed to associate blocks of markers properly across gaps within linkage groups. This was because it depends upon the rankings of recombination fractions at individual markers, and is susceptible to sampling error, typing error and joint selection involving the terminal markers of nearly finished linkage groups. Therefore, the current form of the algorithm is useful mainly to improve local marker ordering in linkage groups obtained in other ways. AVAILABILITY: The source code and supplemental data are http://www.iubio.bio.indiana.edu/soft/molbio/qtl/flipper/ CONTACT: ccrane@purdue.edu.

Algorithms↗
PubMed2003

Barrett's esophagus and Barrett's-related dysplasia.

Barrett's esophagus is a complication of chronic gastroesophageal reflux disease and can be diagnosed when there is an endoscopic abnormality in which a biopsy shows evidence of specialized columnar epithelium, characterized by the presence of acid mucin-containing goblet cells. Much of the controversy in this body of literature relates to the complex anatomy of the esophagogastric junction and the difficulty in precisely identifying this landmark at endoscopy. By definition, in Barrett's esophagus, the squamocolumnar junction is proximal to the esophagogastric junction. Although fundic-type or cardiac-type (junctional) columnar epithelium may be present in Barrett's esophagus, it is only the presence of specialized columnar epithelium that is diagnostic of this condition. Patients with Barrett's esophagus are at risk of progressing to esophageal dysplasia and adenocarcinoma. There are several problems with using dysplasia as a marker for increased cancer risk in these patients, including problems with sampling error and intra- and interobserver variation in the recognition of dysplasia. It may be difficult to distinguish regenerative epithelial changes from dysplasia, low-grade from high-grade dysplasia, and high-grade dysplasia from intramucosal adenocarcinoma. Finally, there are relatively few prospective data evaluating the natural history of high-grade dysplasia. The management of patients with Barrett's-related dysplasia is controversial and varies from institution to institution. Future emphasis should be on cost-effective techniques for sampling as much of the esophageal mucosa as possible in patients who are at the highest risk of progressing to dysplasia and adenocarcinoma. Identification of biomarkers that identify such patients before the histologic recognition of dysplasia will be an area of intensive research.

Barrett Esophagus↗
PubMed2006

Temporal artery biopsy for diagnosing giant cell arteritis: the longer, the better?

OBJECTIVE: To investigate the relation between temporal artery biopsy (TAB) length and diagnostic sensitivity for giant cell arteritis. METHODS: Histological TAB reports generated from four hospital pathology departments were reviewed for demographics, histological findings, and formalin fixed TAB lengths. A biopsy was considered positive for giant cell arteritis if there was a mononuclear cell infiltrate predominating at the media-intima junction or in the media. RESULTS: Among 1821 TAB reports reviewed, 287 (15.8%) were excluded because of missing data, sampling errors, or age < 50 years. Mean TAB length of the 1520 datasets finally analysed (67.2% women; mean (SD) age, 73.1 (10.0) years) was 1.33 (0.73) cm. Histological evidence of giant cell arteritis was found in 223 specimens (14.7%), among which 164 (73.5%) contained giant cells. Statistical analyses, including piecewise logistic regression, identified 0.5 cm as the TAB length change point for diagnostic sensitivity. Compared with TAB length of < 0.5 cm, the respective odds ratios for positive TAB without and with multinucleated giant cells in samples > or = 0.5 cm long were 5.7 (95% confidence interval, 1.4 to 23.6) and 4.0 (0.97 to 16.5). CONCLUSIONS: A fixed TAB length of at least 0.5 cm could be sufficient to make a histological diagnosis of giant cell arteritis.

Aged↗
PubMed1990

Distinguishing error from chaos in ecological time series.

Over the years, there has been much discussion about the relative importance of environmental and biological factors in regulating natural populations. Often it is thought that environmental factors are associated with stochastic fluctuations in population density, and biological ones with deterministic regulation. We revisit these ideas in the light of recent work on chaos and nonlinear systems. We show that completely deterministic regulatory factors can lead to apparently random fluctuations in population density, and we then develop a new method (that can be applied to limited data sets) to make practical distinctions between apparently noisy dynamics produced by low-dimensional chaos and population variation that in fact derives from random (high-dimensional) noise, such as environmental stochasticity or sampling error. To show its practical use, the method is first applied to models where the dynamics are known. We then apply the method to several sets of real data, including newly analysed data on the incidence of measles in the United Kingdom. Here the additional problems of secular trends and spatial effects are explored. In particular, we find that on a city-by-city scale measles exhibits low-dimensional chaos (as has previously been found for measles in New York City), whereas on a larger, country-wide scale the dynamics appear as a noisy two-year cycle. In addition to shedding light on the basic dynamics of some nonlinear biological systems, this work dramatizes how the scale on which data is collected and analysed can affect the conclusions drawn.

Chickenpox↗
PubMed2003

Development of a flow cytometric method to determine DNA ploidy of oesophageal cancer cells obtained by forceps biopsy samples during oesophago-gastro-duodenoscopy.

BACKGROUND: The DNA content of oesophageal tumour cells is a prognostic factor in untreated patients. To investigate whether DNA ploidy is useful to select patients for neoadjuvant therapy it is of interest to develop a method allowing reliable flow cytometric analysis of the DNA content of tumour cells obtained by forceps biopsy during endoscopy before start of therapy. METHODS: Freshly frozen forceps biopsy samples from 30 patients with oesophageal cancer were disaggregated. DNA was stained with propidium iodide and ploidy was determined by flow cytometry. To enhance sensitivity epithelial cells were simultaneously labelled with anti-cytokeratin antibodies. Results were compared with image analysis. To evaluate the sampling error, parallel measurements were done in 10 patients by image analysis on forceps biopsies obtained during endoscopy before surgery and on the resected tumour. RESULTS: The sensitivity to detect aneuploidy was lower for standard flow cytometry than for image analysis (13 versus 33%). The overall sensitivities were identical using a double labelling technique with additional cytokeratin-staining of the epithelial cells, but divergent results were obtained in 2 cases, where detection of aneuploidy was either possible with image analysis or with double labelling flow cytometry only. DNA content of samples gained by forceps biopsies and surgically resected tumours was concordant in 8 of 10 cases. In 2 patients, aneuploidy was detected only in the surgically resected tumour but not in the pre-operatively obtained forceps biopsies. CONCLUSIONS: A flow cytometric method for routine determination of the DNA ploidy of cells obtained by forceps biopsies from patients with oesophageal cancer was developed and evaluated against image analysis. The technique allows the prediction of DNA content before tumour resection, and might be used for optimising therapy and the patient's quality of live.

Adenocarcinoma↗
PubMed2004

Multiregional sampling reveals a homogenous distribution of Ki-67 proliferation rate in pituitary adenomas.

Ki-67 antigen is used as a marker of proliferative activity that is linked to growth rate, invasiveness and prognosis of pituitary adenomas. So far the distribution of Ki-67 index within an individual adenoma has not been investigated. If Ki-67 antigen expression differs significantly within an individual pituitary adenoma, a sampling error may result when assessing small fragments of adenoma tissue. Such a potential error would diminish the value of Ki-67 as a tool for postoperative patient management considerations. The aim of the present study was to assess Ki-67 proliferation rates in different regions of pituitary adenomas and to statistically analyse these data for potential regional differences within each tumor. Ki-67 proliferation index was assessed in smear preparations of 100 specimens of 26 consecutive patients operated on for pituitary adenoma in the Department of Neurosurgery, Medical University Vienna. Depending on the size and extent of the tumor, a mean of 4 tissue samples (range 2-8) was selected intraoperatively from each adenoma from endosellar, suprasellar, parasellar, and basal sellar dural locations. Overall mean cell proliferation rate measured by Ki-67 was 1.81 +/- 0.90% (range 0.33-3.43%). Histologically invasive adenomas had significantly higher mean Ki-67 proliferation index in all samples from the same tumor than non-invasive adenomas (2.01 +/- 0.91% vs. 1.11 +/- 0.59%; P = 0.024). Multiregional sampling revealed a homogenous distribution of Ki-67 index throughout an individual adenoma with no significant differences between any two different regions on t-test. Our data confirm that location of a biopsy does not influence Ki-67 index. Therefore, Ki-67 index of a single biopsy is representative for the whole individual adenoma. Thus Ki-67 index can be considered a reliable parameter for assessment of cell proliferation rate in adenoma biopsies and may be used for postoperative patient management considerations.

Adenoma↗
PubMed1994

Estimating confidence limits on a standardised mortality ratio when the expected number is not error free.

OBJECTIVE: The aim was to demonstrate how the beta distribution may be used to find confidence limits on a standardised mortality ratio (SMR) when the expected number of events is subject to random variation and to compare these limits with those obtained with the standard exact approach used for SMRs and with a Fieller-based confidence interval. DESIGN: The relationship of the binomial and the beta distributions is explained. For cohort studies in which deaths are counted in exposed and unexposed groups exact confidence limits on the relative risk are found conditional on the total number of observed deaths. A similar method for the SMR is justified by analogy between the SMR and the relative risk found from such cohort studies, and the fact that the relevant (beta) distribution does not require integer parameters. SOURCE OF DATA: Illustrative examples of hypothetical data were used, together with a MINITAB macro (see appendix) to perform the calculations. MAIN RESULTS: Exact confidence intervals that include error in the expected number are much wider than those found with the standard exact method. Fieller intervals are comparable with the new exact method provided the observed and expected numbers (taken to be means of Poisson variates) are large enough to approximate normality. As the expected number is increased, the standard method gives results closer to the new method, but may still lead to different conclusions even with as many as 100 expected. CONCLUSIONS: If there is reason to suppose the expected number of deaths in an SMR is subject to sampling error (because of imprecisely estimated rates in the standard population) then exact confidence limits should be found by the methods described here, or approximate Fieller-based limits provided enough events are observed and expected to approximate normality.

Adolescent↗
PubMed2003

Pineal tumors and associated lesions: the effect of ethnicity on tumor type and treatment.

Pineal region lesions consist of a wide variety of rare tumor types, including deep midline cysts, intrinsic pineal tumors, germ cell tumors and vascular lesions. Advances in microsurgical, endoscopic and stereotactic techniques have helped to lower morbidity and mortality in the care of patients harboring these lesions. Surgery can be the definitive treatment in cysts and benign lesions. This report summarizes the retrospective experience of the authors with 64 pineal region and associated lesions encountered in multiple institutions over the last 20 years. Histology was obtained in 53 out of 64 radiographically apparent lesions. Direct surgical biopsy of solid and vascular tumors in the pineal region enables precise histological assessment of mixed tumors. By avoiding sampling error, precise treatment can be planned. This series, along with previously published data, shows a much higher incidence of intrinsic pineal tumors, glial tumors and nongerminomatous germ cell tumors in series from North America and Europe than in those from Japan and Korea, where germinoma is much more common. We experienced an incidence of 20.4% germinoma out of 49 solid and vascular pineal tumors, while other authors have described incidences of 51.2 and 53.5%, respectively. The fact that histology is more diverse in Western populations leads to a need to have more representative sampling. Early surgical resection combined with diversion of cerebrospinal fluid is effective in the treatment of pineal lesions and seems to be superior to the alternative of treatment based on the diagnostic response to radiation and/or on tumor markers alone.

Adolescent↗
PubMed1993

Copia-like retrotransposable element evolution in diploid and polyploid cotton (Gossypium L.).

Copia-like retrotransposable elements were identified in allotetraploid cotton, Gossypium hirsutum, and two species representing its diploid progenitors, G. herbaceum and G. raimondii. These elements are present in high copy number in all three species. Because the two diploid genomic groups have been isolated on opposite sides of the world for 6-11 million years, horizontal transfer of elements between these species is highly unlikely. Elements were intensively sampled to generate a model of copia-like retrotransposable element evolution in systems where vertical transmission is the sole probable means of descent. Copia-like retrotransposon diversity is equally great in all three Gossypium species. Despite this high heterogeneity, analysis of 89 partial reverse transcriptase sequences resulted in the recognition of nine sharply differentiated retrotransposon lineages, each containing elements that share high sequence similarity. No evidence of horizontal transfer from other taxa was obtained. Phylogenetic analyses demonstrate that element topologies are incongruent with Gossypium phylogeny. Consideration of processes that obscure phylogenetic reconstruction of multigene families (including sampling error, variable degrees of orthology and paralogy, differential lineage age and lineage loss and/or proliferation) demonstrates that incongruence between organismal and retrotransposon trees is expected under conditions in which vertical processes are the sole means of transmission. Identification of closely related elements between species allowed rates of copia-like retrotransposon sequence evolution to be estimated as approximately 10(-9) nucleotide substitutions/site/year. These rates are consistent with the interpretation that these retrotransposons have been evolving under functional constraints for most of the time frame bracketed by the species studied. Extrapolation of these results to previous studies that sampled from more highly divergent taxa indicates that horizontal transfer need not be invoked to explain observed phylogenetic patterns.

Amino Acid Sequence↗
PubMed2003

Single versus double testing of meat-juice samples for Salmonella antibodies, in the Danish pig-herd surveillance programme.

In Denmark, a national serological surveillance-and-control programme for Salmonella in pigs has been in operation since 1995. The programme is based on the Danish mix-ELISA and uses double testing (two ELISA-wells used per sample) of meat-juice samples taken in relation to slaughter. All herds are classified monthly into one of the three levels; the classification is based on the percentage of positive serological results in the previous 3 months. In connection with evaluation of the programme in 2001, we investigated whether single testing (testing in one well only) could be expected to be sufficiently precise compared to double testing. Data from the year 2000 were used, and mathematical modelling. Single testing was simulated by randomised selection of one of the two results in the double testing. A slight increase in the prevalence of Salmonella-positive samples (1.02-1.09 times more through the four quarters of the year 2000) was found in the simulated single testing, as compared to the double testing. Around 0.5% of the herds would be allocated to another herd level in single testing-almost equal numbers one level up and one level down. No herd being seronegative in double testing would be allocated to levels 2 or 3 (herds with >40 or >70%, respectively, serological reactors) in single testing. The prevalence of "false-positive" diagnoses (positive in single testing and negative in double testing) and inversely defined "false-negative" diagnoses varied from 4.2 to 8.7% and from 3.2 to 4.5%, respectively, through the four quarters of the year 2000. The probability of allocating a herd to a wrong level due to sampling error was on the average 6.2 (varying from 1.66 to over 100) times higher than the probability of allocating a herd to a wrong level due to the test inaccuracy introduced by going from double to single testing. This is, however, an average; a herd with a true prevalence close to one of the level border cut-offs (40 and 70% weighted seroprevalence, respectively) would have a higher risk of being allocated to a wrong level than a herd with a true prevalence far from the level border cut-offs. The results are based on the current Danish sample sizes in the surveillance scheme, which implies that 60, 75 or 100 samples are taken annually in a herd, depending on its size. Other sample sizes would produce other results.

Abattoirs↗
PubMed1985

Matched groups analysis method.

The matched group analysis method in prospective studies is a variation of matched pairs analysis with the advantage of sampling error avoided in the choosing of pairs of subjects. An example is shown in which comparisons are given of mortality in triads of cases classified by the tar and nicotine level of the cigarettes they smoked and matched on 10 variables. The method can also be used in retrospective and autopsy studies.

Adult↗
PubMed1998

Intraoperative confirmation of parathyroid tissue during parathyroid exploration: a retrospective evaluation of the frozen section.

Although the frozen section is widely used to identify tissue type during parathyroid exploration in patients with hyperparathyroidism, questions regarding its accuracy have been raised. Frozen section error has been identified as a significant factor contributing to surgical failure. The purpose of this study was to establish the accuracy of frozen section in this setting and to identify pitfalls underlying frozen section error. The final pathologic diagnoses were compared with the paired frozen section diagnoses for all patients who underwent parathyroid exploration at this institution in the period between 1984 and 1997. For those cases in which a discrepancy was identified, the original histopathology slides and the medical records were reviewed. Of the 1579 frozen sections, a definitive and accurate diagnosis could not be determined in 20 cases (1.3%); in 7 (0.4%), the frozen section diagnosis was deferred, and in the other 13 cases, the frozen section diagnosis was incorrect. Overall accuracy rate was 99.2% after deferred cases were excluded. Frozen section artifact, sampling error, and judgmental error contributed to deferred or incorrect diagnoses. Several features confounded the distinction between parathyroid and thyroid tissue in 10 cases: the coexistence of parathyroid and nodular thyroid disease; intrathyroidal parathyroid glands showing conspicuous follicle formations or abundant oncocytic cells; and thyroid nodules with fatty stroma. The frozen section is a highly reliable means of identifying tissue type during parathyroid exploration. In exceptional cases, however, the distinction between parathyroid and thyroid tissue may not be possible owing to a striking overlap seen at the clinical, gross, and microscopic levels.

Artifacts↗