Early identification of patients at low risk for poor outcome after acute upper gastrointestinal hemorrhage would allow reduction of diagnostic and therapeutic interventions. We identified six early predictors of good outcome: age less than 75 years, no unstable comorbid illness, no ascites found on physical examination, normal prothrombin time, and, within an hour after presentation, systolic blood pressure of 100 mm Hg or greater and nasogastric aspirate free of fresh blood. Presence of all six predictors defined the low-risk population. Among 162 patients in the development and retrospective validation phases of our study, all 74 low-risk patients had good outcomes. A prospective validation study of 111 patients further established the accuracy of our predictive method; only two of 52 low-risk patients had poor outcomes. Application of our method should allow more selective management of patients with acute upper gastrointestinal hemorrhage.

Two adult male patients with acute leukemia developed a fatal Budd-Chiari-like illness while receiving 6-thioguanine. Both had previously received cytosine arabinoside. Antemortem and postmortem specimens of liver showed changes characteristic of toxic veno-occlusive disease. Similar findings have been described after ingestion of certain plant alkaloids and after treatment with arsphenamine, urethane, and ionizing radiation to the liver. We are unaware of any published reports of veno-occlusive disease of the liver after treatment with either 6-thioguanine or cytosine arabinoside. Although 6-thioguanine was most likely responsible for this syndrome, it is not possible to eliminate cytosine arabinoside as the causative agent. Since both drugs are occasionally used for benign conditions, physicians should be aware of this possible complication.

No abstract

We have assessed long-term trends in the use of laboratory tests among patients hospitalized on the medical service of a large teaching hospital. A significant decline in numbers of chemistry tests and no growth in numbers of microbiology or hematology tests or roentgenograms were noted between 1970 and 1977. Per-patient increases in laboratory costs and charges were considerably less than increases in total hospitalization costs and charges. These findings contrast to an average increase of 13.8% per year in numbers of laboratory tests for hospital laboratories in this country between 1970 and 1975. Factors responsible for the observed stability in numbers of laboratory tests per patient per hospitalization included, but were not limited to, increased reliance on test batteries as opposed to individual tests; changing patterns of care leading to decreased use of nonautomated tests; and the impact of administrative and educational strategies directed toward optimum use of the laboratory.

Studies often suggest that accepted clinical predictors actually have little predictive strength. One explanation for some such results is the presence of workup bias. To explore the effects of workup bias in prediction research, the authors modeled the effects of workup bias on the ability of early clinical findings to predict intracerebral hemorrhage in patients with stroke. In a simulated biased sample, workup bias resulted in distorted operating characteristics for those clinical findings influencing application of the "gold standard" and for other related findings. Sensitivity was increased, but both specificity and likelihood ratios were decreased in the biased sample. Workup bias can spuriously decrease predictive abilities for accepted clinical findings when such findings guide application of the "gold standard." Investigators should be aware of the potential effects of workup bias, search for clues to its occurrence, and interpret study results carefully when it is present.