Overall Level Of Agreement Research Articles

Uso apropiado de los antiinflamatorios no esteroideos en reumatología: documento de consenso de la Sociedad Española de Reumatología y el Colegio Mexicano de Reumatología

Objective To develop guidelines for the appropriate use of NSAIDs in rheumatology. Methods We used a methodology modified from the one developed by RAND/UCLA. Two groups of panellists were selected, one by the CMR and another by the SER. Recommendations were proposed from nominal groups and the agreement to them was tested among rheumatologists from both societies by a tworound Delphi survey. The analysis of the second Delphi round supported the generation of the final set of recommendations and the assignment of a level of agreement to each of them. Systematic reviews of five recommendations in which the agreement was low or was divided were also carried out. Results Here we present recommendations for the safe use of NSAIDs in rheumatic diseases, based on the best available evidence, expert opinion, the agreement among rheumatologists, and literature review. The trend is to reduce the frequency, duration and dose of NSAIDs in favour of non-pharmacological measures, analgesic drugs or disease modifying drugs. In addition, the recommendations help to identify profiles for increased toxicity, with an emphasis on gastrointestinal and cardiovascular risks. The recommendations deal with the course of action and monitoring in different risk groups and in patients using antiplatelet or anticoagulant drugs. The overall level of agreement is high. Conclusions The NSAIDs are safe and effective drugs for the treatment of rheumatic diseases. However, it is necessary to individualize its use according to their risk profile.

Appropriate Use of Non-steroidal Anti-inflammatory Drugs in Rheumatology: Guidelines From the Spanish Society of Rheumatology and the Mexican College of Rheumatology

ObjectiveTo develop guidelines for the appropriate use of NSAIDs in rheumatology. MethodsWe used a methodology modified from the one developed by RAND/UCLA. Two groups of panellists were selected, one by the CMR and another by the SER. Recommendations were proposed from nominal groups and the agreement to them was tested among rheumatologists from both societies by a 2-round Delphi survey. The analysis of the second Delphi round supported the generation of the final set of recommendations and the assignment of a level of agreement to each of them. Systematic reviews of 5 recommendations in which the agreement was low or was divided were also carried out. ResultsHere we present recommendations for the safe use of NSAIDs in rheumatic diseases, based on the best available evidence, expert opinion, the agreement among rheumatologists, and literature review. The trend is to reduce the frequency, duration, and dose of NSAIDs in favour of non-pharmacological measures, analgesic drugs, or disease modifying drugs. In addition, the recommendations help to identify profiles for increased toxicity, with an emphasis on gastrointestinal and cardiovascular risks. The recommendations deal with the course of action and monitoring in different risk groups and in patients using antiplatelet or anticoagulant drugs. The overall level of agreement is high. ConclusionsThe NSAIDs are safe and effective drugs for the treatment of rheumatic diseases. However, it is necessary to individualize its use according to their risk profile.

Use of the GMFCS in infants with CP: the need for reclassification at age 2 years or older

The stability of the Gross Motor Function Classification System (GMFCS) over time is described in 77 infants (41 boys, 36 girls) with cerebral palsy (CP; mean age 19.4mo [SD 1.6 mo]; 27 unilateral spastic, 42 bilateral spastic, eight dyskinetic type) and in the same children at follow-up at age 2 to 4 years. The overall level of agreement over time (linear weighted kappa) was 0.70 (95% confidence interval [CI] 0.61-0.79). The overall percentage of children whose GMFCS level changed one or two levels was 42%, of which the majority were reclassified to a less functional level (McNemar's Chi(2) test p=0.11). The chance that children initially classified in the combination of GMFCS Levels I, II, and III would subsequently be classified in the same level in early childhood was 96% (positive predictive value [PPV] 0.96, 95% CI 0.85-0.99), whereas the PPV for the combination of Levels I and II was 0.88, 95% CI 0.70-0.96. These findings indicate that GMFCS classification in infants is less precise than classification over time in older children. In conclusion, children can be classified by the GMFCS early on, but there is a need for reclassification at age 2 or older as more clinical information becomes available.

Accuracy of chest radiograph interpretation by emergency physicians

Chest radiographs are one of the most complex imaging modalities to interpret. The objective of this study was to assess how accurately emergency physicians interpreted chest radiographs in relation to radiologist reports. Radiological descriptions of chest radiographs from 667 emergency department (ED) patients, aged 14 to 84 years, were retrospectively reviewed. The overall level of agreement and agreement for different categories (congestive heart failure, pneumonia, and other) were assessed in 312 cases for which reports were available. Statistical analysis was done using the chi-square test, and interobserver agreement was calculated. The proportion of agreement between emergency physician and radiologist reports for normal, congestive heart failure, and pneumonia cases was 84.3%, 41.4%, and 41.4%, respectively. Emergency department physicians frequently missed specific radiographic abnormalities, and there was considerable discrepancy between their interpretations and those of trained radiologists. The agreement for some diagnostic categories such as pneumonia and congestive heart failure was low. This study's findings emphasize the need for improving interpretive skills among ED physicians.

Prospective Evaluation of the Clinical Application of the American College of Cardiology Foundation/American Society of Echocardiography Appropriateness Criteria for Transthoracic Echocardiography

We sought to prospectively evaluate the clinical application of the American College of Cardiology Foundation/American Society of Echocardiography Appropriateness Criteria (AC) for transthoracic echocardiography in a single-center university hospital. Indications for transthoracic echocardiograms (TTE) were prospectively determined for consecutive studies by 2 reviewers and categorized, according to the AC for TTE, as appropriate (A) or inappropriate (I). The overall level of agreement in characterizing appropriateness between reviewers was high (kappa = 0.83). Among the 1,553 studies for which a primary indication was determined, 89% were covered in the AC for TTE. Of these studies, 89% were A, and 11% were I. New important TTE abnormalities were more common on A compared with I studies (40% vs. 17%, p < 0.001), and noncardiac specialists more frequently ordered I studies (13% vs. 9%, p = 0.04). In conclusion, the AC for TTE encompasses the majority of clinical indications for TTE and appears to reasonably stratify TTE ordering. However, revisions will be needed to fully capture and stratify appropriate clinical practice.

Interobserver variation in the classification of thymic tumours – a multicentre study using the WHO classification system

To test the reproducibility of the current World Health Organization (WHO) classification of thymic epithelial tumours and to determine the level of interobserver variation within a group of pathologists, all with experience and expertise in thoracic pathology. Ninety-five thymic tumours were circulated to a group of 17 pathologists in the UK and The Netherlands over a 1-year period. Participants were asked to classify them according to WHO criteria. The diagnoses were subjected to statistical analysis and kappa values calculated. The overall level of agreement was moderate (kappa 0.45). When the categories were reduced in number by creating two groups, (A + AB + B1 + B2 and B3 + C), the level of agreement increased to 0.62. An alternative grouping (A + AB + B1 and B2 + B3 + C) increased it slightly further. The best agreement was in tumour types A and AB. Difficulties arose in distinguishing B1 tumours from B2 tumours and B2 tumours from B3 tumours. Although the WHO system describes a number of well-defined tumour types with clear diagnostic criteria, the overall level of agreement was moderate and improved if some groups were amalgamated.

The Impact of Hospital-Acquired Infection on Outcome in Acute Pancreatitis

Little is known regarding the impact of hospital-acquired infection (HAI) in acute pancreatitis (AP). We conducted a population-based assessment of the impact of HAI on outcome in AP. Patient data were obtained from the Cardinal Health Clinical Outcomes Research Database, a large population-based data set. Cases with principal diagnosis by International Classification of Diseases, ninth revision, clinical modification 577.0 (AP) between January 2004 and January 2005 were identified. These cases were linked with recently reported HAI data collected by the Pennsylvania Health Care Cost Containment Council. Identification of HAI was based on definitions set forth by the National Nosocomial Infection Surveillance System. We conducted a 5:1 multivariate propensity-matched cohort study to determine the independent contribution of HAI to in-hospital mortality, length of stay (LOS), and hospital charges. From 177 participating hospitals, there were 11,046 AP cases identified. Eighty-two (0.7%) patients developed an HAI. Mortality in the overall AP population was 1.2% vs 11.4% among 405 matched non-HAI controls vs 28.4% among patients who developed HAI (chi(2) test, P < .0001). Fifteen percent of all deaths was associated with an HAI. Both average LOS and hospital charges were significantly increased among patients with HAI compared with matched non-HAI controls. We determined that HAI had a major impact on mortality in AP. Patients who developed HAI also had significantly increased LOS and hospital charges. These differences were not explained by increased disease severity alone. Reducing HAI is an important step to improving outcome in AP.

Application of Appropriateness Criteria to Stress Single-Photon Emission Computed Tomography Sestamibi Studies and Stress Echocardiograms in an Academic Medical Center

The purpose of this study was to apply published appropriateness criteria for single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) in a single academic medical center. The American College of Cardiology Foundation (ACCF) and the American Society of Nuclear Cardiology (ASNC) have developed appropriateness criteria for stress SPECT MPI to address concern about the growth in cardiac imaging studies. We retrospectively examined 284 patients who underwent stress SPECT MPI and 298 patients who underwent stress echocardiography before publication of these criteria. The overall level of agreement in characterizing appropriateness between 2 experienced cardiovascular nurse abstractors was modest (kappa = 0.56), but noticeably poorer (kappa = 0.27) for patients with previous SPECT or echo studies. Similar percentages of each imaging modality were assigned to the 3 appropriateness categories: 64% of stress SPECT and 64% of stress echo studies were classified appropriate; 11% of stress SPECT and 9% of stress echo were of uncertain appropriateness; and 14% of stress SPECT and 18% of stress echo were inappropriate. Of the inappropriate studies, 88% were performed for 1 of 4 indications. Approximately 10% of the patients were unclassifiable. Application of existing SPECT MPI appropriateness criteria is demanding and requires an established database or detailed data collection, as well as a number of assumptions. Fourteen percent of stress SPECT studies and 18% of stress echo studies were performed for inappropriate reasons. Quality improvement efforts directed at reducing the number of these inappropriate studies may improve efficiency in the health care system.

Pilot Study of Prevalence of High-Risk Human Papillomavirus Genotypes in Israeli Jewish Women Referred for Colposcopic Examination

The objective of the present study was to determine the prevalence of high-risk (HR) human papillomavirus (HPV) genotypes in a group of Israeli Jewish women referred for colposcopic examination. Scrape specimens were prospectively collected from 84 women referred for colposcopic examination. All the women underwent Papanicolaou (Pap) smears and colposcopies, and some also underwent cervical or loop electrosurgical excision procedure biopsy. HR HPV was detected in scrape specimens (Amplicor HPV test; Roche Molecular Systems), and the individual genotypes in these specimens were identified (HPV GenoArray test kit; Hybribio Ltd., Hong Kong). Forty-one (49%) specimens were positive by the Amplicor HPV test. Sixty-four samples (41 positive and 23 negative by the Amplicor HPV test) were also assayed by use of the HPV GenoArray kit. The overall level of agreement between the two assays was 93.8% (Cohen's kappa = 0.98). HR genotypes were found in 37/41 (90%) HPV-positive samples. The prevalences of the HR HPV genotypes in the 37 HPV-positive samples were 41% of patients for HPV type 16 (HPV-16), 22% for HPV-39, 19% for HPV-52, and 14% for HPV-18. Forty-one percent of these patients were infected with a single HR genotype, whereas 59% were infected with mixtures of HR genotypes. The presence of a relatively high percentage of HPV types 39 and 52 and the relatively high incidence of infections with mixtures of genotypes may be one of the reasons for the low rate of conversion from high-grade squamous intraepithelial lesions to invasive carcinoma in Israeli women. Larger and more comprehensive studies are warranted to investigate this issue in greater detail.

The Alberta Breakthrough Pain Assessment Tool for Cancer Patients: A Validation Study Using a Delphi Process and Patient Think-Aloud Interviews

Breakthrough pain is a prevalent cancer pain syndrome, and research is needed to identify more effective interventions to manage it. A validated tool to assess breakthrough pain in a standard and reliable manner is urgently needed to support the conduct of clinical trials in breakthrough pain. To address this need, we developed a breakthrough pain assessment tool for research purposes. The current study was undertaken to gather validity evidence for this breakthrough pain assessment tool, using a Delphi process involving an expert panel review, followed by a think-aloud process involving patients with cancer-related breakthrough pain. Two expert panels were formed: a national panel (within Canada; n = 16) and an international panel (including experts from North America, UK, Europe, the Middle East, Australia, and New Zealand; n = 22). Each panel participated in one anonymous survey round. Response rates were 56% (national panel) and 73% (international panel). The Delphi process revealed substantial consensus on the content of the tool, which increased between rounds of review. The overall level of agreement with the tool, averaged over the four evaluated aspects of all items, was 80% among national panelists and 88% among international panelists. Nine patients completed the think-aloud study. They were able to understand and complete the tool and provided specific direction on its improvement. The validity evidence gathered in this study suggests the Alberta Breakthrough Pain Assessment Tool is conceptually grounded and is understandable by patients and clinicians. Further validation of this tool as an assessment measure within clinical trials research is warranted.

Multicenter Laboratory Evaluation of the MB/BacT Mycobacterium Detection System and the BACTEC MGIT 960 System in Comparison with the BACTEC 460TB System for Susceptibility Testing of Mycobacterium tuberculosis

In this multicenter study, the reliability of two nonradiometric, fully automated systems, the MB/BacT and BACTEC MGIT 960 systems, for testing the susceptibilities of 82 Mycobacterium tuberculosis strains to isoniazid, rifampin, ethambutol, and streptomycin was evaluated in comparison with the radiometric BACTEC 460TB system. The arbitration of discrepant results was done by the reanalysis of the strain, the determination of the MIC, and the molecular characterization of some resistance determinants. The overall level of agreement with BACTEC 460TB results was 96% with the MB/BacT test and 97.2% with the BACTEC MGIT 960 system. With both methods, the level of agreement with BACTEC 460TB results was 96.3% for isoniazid, 98.8% for rifampin, and 98.8% for ethambutol. The level of agreement for streptomycin was 90.2% with MB/BacT and 97.5% with BACTEC MGIT 960. Overall, there were 11 very major errors and 2 major errors with the MB/BacT method and 5 very major errors and 2 major errors with the BACTEC MGIT 960 system. In general, the MB/BacT and BACTEC MGIT 960 systems showed good performance for susceptibility testing with first-line antituberculosis drugs.

A Comparison of Self-Reported Use of Behavioral Health Services With Medicaid Agency Records in Minnesota

A Comparison of Self-Reported Use of Behavioral Health Services With Medicaid Agency Records in Minnesota

Evaluation of the BD Phoenix Automated Microbiology System for Identification and Antimicrobial Susceptibility Testing of Enterobacteriaceae

We evaluated the accuracy of the BD Phoenix system for the identification (ID) and antimicrobial susceptibility testing (AST) of 251 isolates of the family Enterobacteriaceae representing 31 species. Organisms were inoculated onto the Phoenix panel according to the manufacturer's instructions. The results from conventional biochemical tests were used for the reference method for ID. Agar dilution, performed according to the CLSI guidelines, was the reference AST method. Essential and categorical agreements were determined. The overall levels of agreement for the genus- and species-level identifications were 95.6% and 94.4%, respectively. Fourteen isolates were incorrectly identified by the Phoenix system; 10 of these were incorrectly identified to the species level. Three of these were Enterobacter (Pantoea) species and four of these were Shigella spp. misidentified as Escherichia coli. For AST results, the essential and categorical agreements were 98.7% and 97.9%, respectively. The very major error, major error, and minor error rates were 0.38%, 0.33%, and 1.8%, respectively. Six isolates (three E. coli isolates and three Klebsiella isolates) were extended-spectrum beta-lactamase producers. All six were flagged by the Phoenix system expert rules. The Phoenix system compares favorably to traditional methods for ID and AST of Enterobacteriaceae.

Detecting errors in English article usage by non-native speakers

One of the most difficult challenges faced by non-native speakers of English is mastering the system of English articles. We trained a maximum entropy classifier to select among a/an, the, or zero article for noun phrases (NPs), based on a set of features extracted from the local context of each. When the classifier was trained on 6 million NPs, its performance on published text was about 83% correct. We then used the classifier to detect article errors in the TOEFL essays of native speakers of Chinese, Japanese, and Russian. These writers made such errors in about one out of every eight NPs, or almost once in every three sentences. The classifier's agreement with human annotators was 85% (kappa = 0.48) when it selected among a/an, the, or zero article. Agreement was 89% (kappa = 0.56) when it made a binary (yes/no) decision about whether the NP should have an article. Even with these levels of overall agreement, precision and recall in error detection were only 0.52 and 0.80, respectively. However, when the classifier was allowed to skip cases where its confidence was low, precision rose to 0.90, with 0.40 recall. Additional improvements in performance may require features that reflect general knowledge to handle phenomena such as indirect prior reference. In August 2005, the classifier was deployed as a component of Educational Testing Service's Criterion$^{SM}$ Online Writing Evaluation Service.

Real‐time perfusion adenosine stress echocardiography versus myocardial perfusion adenosine scintigraphy for the detection of myocardial ischaemia in patients with stable coronary artery disease

Real-time perfusion (RTP) contrast echocardiography using low mechanical index power modulation technique allows for simultaneous myocardial perfusion and wall motion analysis. RTP-adenosine stress echocardiography (ASE) could be an alternative to dobutamine-atropine stress echocardiography; more tolerable for the patients and possibly similarly accurate. We aimed to evaluate RTP-ASE for the detection of myocardial ischaemia, compared to 99mTc-sestamibi single-photon emission computed tomography (SPECT). Patients with suspected coronary artery disease, admitted to SPECT evaluation, were prospectively invited to participate. Patients underwent RTP imaging (SONOS 5500) using infusion of Sonovue (Bracco, Milano, Italy) before and during ASE. Two separate readers performed off-line analysis of myocardial perfusion and wall motion by RTP-ASE. A perfusion defect was the principal marker of ischaemia. Wall motion assessment was used to evaluate ischaemia in segments with perfusion artefacts. Each segment was attributed to one of the three main coronary vessel areas of interest: the left anterior descending (LAD); the left circumflex (LCx) and the right posterior descending (RPD). Normal SPECT at stress was judged normal at rest. In 33 patients, 99 coronary territories were analysed by SPECT and RTP-ASE. SPECT showed evidence of ischaemia in 9 of 33 patients. For the detection of ischaemia, the overall level of agreement between RTP-ASE and SPECT was 92% in all segments. The level of agreement was 88% in LAD, 97% in LCx and 91% in RPD segments. Real-time perfusion-adenosine stress echocardiography using power modulation could be an accurate and feasible tool for evaluation of ischaemia in patients with suspected coronary artery disease. The results from this study need confirmation by a study of a larger patient sample.

A Direct Comparison of Drug Susceptibility to HIV Type 1 from Antiretroviral Experienced Subjects as Assessed by the Antivirogram and PhenoSense Assays and by Seven Resistance Algorithms

HIV-1 drug resistance methodologies are being increasingly utilized to guide treatment decisions; however, information comparing the various assays is limited. Duplicate plasma samples from 70 ART-experienced subjects were analyzed by both the Antivirogram and PhenoSense phenotypic assays and the results compared. HIV genotypes were also obtained and analyzed using seven different resistance algorithms. These results were also compared with the phenotypic assay results. Concordances between the phenotypic tests and between each algorithm, and between the two phenotypic assays were calculated and kappa coefficients (KC) determined. Overall agreement between the two phenotypic assays was good (86.9% concordance; KC 0.621). The highest concordance by drug class was seen for protease inhibitors (93.4%; KC 0.679) and the lowest (79.8%; KC 0.549) for nucleoside reverse transcriptase inhibitors. Concordance between the two phenotypic assays, when evaluating individual drugs, was good for all drugs tested except for abacavir, zalcitabine, and indinavir. Agreement between the seven algorithms and each phenotypic assay was variable, though most had good or excellent agreement. The highest overall level of agreement for an individual drug was observed when comparing lamivudine susceptibility to either assay. Concordance for abacavir, didanosine, zalcitabine, and saquinavir was generally problematic when comparing one or more phenotypic assays to the drug resistance predictive algorithms. In conclusion, results comparing these two phenotypic tests were mostly similar, but comparisons of the predictive resistance algorithms for specific drugs, as well as to specific phenotypic assays, were more inconsistent.

Comparison of CT Duodeno-Cholangiopancreatography to ERCP for Assessing Biliary Obstruction

The authors have developed a CT protocol, CT duodeno-cholangiopancreatography (CDCP), which is performed during a single contrast-enhanced phase, proceeding cranially, allowing enhancement of the pancreas during its parenchymal phase followed by enhancement of the liver during its portal-venous phase. This retrospective pilot study evaluates CDCP compared with endoscopic retrograde cholangiopancreatography (ERCP) as a diagnostic tool for assessing the cause and level of biliary obstruction. Forty-one patients with jaundice underwent CDCP and ERCP between October 2002 and May 2004. Pathologic confirmation was obtained in 31 of the 41 (76%) patients. The sensitivity, specificity, and kappa values of CDCP and ERCP compared with pathology were calculated for tumors and stones. Pathology-proven cases included 7 cases of stones, 23 tumors, and 1 other cause of obstruction. The overall level of agreement of diagnoses between CDCP and pathology was 29 of 31 (93.5%); that between CDCP and ERCP was 36 of 41 (88%). Comparing CDCP to pathology for tumors, the sensitivity was 100%, the specificity was 89%, and the kappa was 0.92 (95% CI 0.76-1.0). For stone detection, CDCP had a sensitivity of 86%, a specificity of 100%, and a kappa value of 0.90 (95% CI 0.72-1.0). For level of obstruction of the common bile duct, comparing CDCP to ERCP, observations agreed in 31 of the 36 (86%) cases; for the pancreatic duct, observations agreed in 24 of the 25 (96%) cases. CDCP is a noninvasive diagnostic tool that can be used to assess the cause and level of obstruction. A blinded prospective study would be valuable to further assess the merits of CDCP.

Comparing predicted and observed spatial boundaries of geologic phenomena: Automated Proximity and Conformity Analysis applied to ice sheet reconstructions

Comparing predicted with observed geologic data is a central element of many aspects of research in the geosciences, e.g., comparing numerical ice sheet models with geomorphic data to test ice sheet model parameters and accuracy. However, the ability to verify predictions using empirical data has been limited by the lack of objective techniques that provide systematic comparison and statistical assessment of the goodness of correspondence between predictions of spatial and temporal patterns of geologic phenomena and the field evidence. Much of this problem arises from the inability to quantify the level of agreement between straight or curvilinear features, such as between the modeled extent of some geologic phenomenon and the field evidence for the extent of the phenomenon. Automated Proximity and Conformity Analysis (APCA) addresses this challenge using a system of Geographic Information System-based buffering that determines the general proximity and parallel conformity between linear features. APCA results indicate which modeled output fits empirical data, based on the distance and angle between features. As a result, various model outputs can be sorted according to overall level of agreement by comparison with one or multiple features from field evidence, based on proximity and conformity values. In an example application drawn from glacial geomorphology, APCA is integrated into an overall model verification process that includes matching modeled ice sheets to known marginal positions and ice flow directions, among other parameters. APCA is not limited to ice sheet or glacier models, but can be applied to many geoscience areas where the extent or geometry of modeled results need to be compared against field observations, such as debris flows, tsunami run-out, lava flows, or flood extents.

Patients’ and nurses’ perceptions of nursing problems in an acute care setting

This paper reports a study to determine the degree of agreement or disagreement between nurses and patients in their perceptions of the presence, severity, and importance of nursing problems. Patient experiences, values and preferences are increasingly acknowledged as important factors underpinning healthcare decision-making. The ability to identify patient problems accurately is an important prerequisite for planning and implementing individualized high quality care. A convenience sample of patients (n = 80) and Registered Nurses (n = 30) in an acute care setting responded to a 43-item questionnaire. Findings. Nurses identified patients' problems with a sensitivity of 0.53 and a positive predictive value of 0.50. Patients identified several severe problems that were not identified by nurses, particularly problems with nutrition, sleep, pain, and emotions/spirituality. Nurses underestimated the severity in 47% of mutually-identified problems. An overall level of agreement of 44% was found on the importance of patient problems. Low levels of agreement on severity and importance were related more to individual differences than to systematic differences. Nurses need to be more aware that patients and nurses often hold disparate views of the priorities in nursing care. To plan individualized nursing care effectively, nurses need to elicit and use individual patients' preferences more systematically in care planning.

A comparative study of three-dimensional saline infusion sonohysterography and diagnostic hysteroscopy for the classification of submucous fibroids

The purpose of this study was to compare three-dimensional saline infusion sonohysterography (3D SIS) and diagnostic hysteroscopy for the diagnosis and classification of submucous uterine fibroids. This was a prospective double-blind study of 49 women who presented with a history of menorrhagia, diagnosed on non-enhanced two-dimensional ultrasonography with submucous fibroids. Fibroids were classified on 3D SIS according to the proportion of fibroid contained within the endometrial cavity, using the European Society of Hysteroscopy Classification of Submucous Fibroids. These results were then compared with the findings at diagnostic hysteroscopy. A total of 61 submucous fibroids was identified in 49 symptomatic women. Diagnostic hysteroscopy confirmed these findings in all cases. There was agreement between the two methods in 11/12 cases of Type 0 fibroids (92%), 34/37 (92%) of Type I fibroids and 9/12 (75%) of Type II fibroids. The overall level of agreement was good with a kappa value of 0.80. There is a good overall agreement between 3D SIS and diagnostic hysteroscopy in classification of submucous fibroids. Agreement is better in cases where a greater proportion of the fibroid is contained within the uterine cavity.