Matched-pair analysis of hematopoietic progenitor cell mobilization using G-CSF vs. cyclophosphamide, etoposide, and G-CSF: enhanced CD34+ cell collections are not necessarily cost-effective.

Using matched-pair analysis, we compared two popular methods of stem cell mobilization in 24 advanced-stage breast cancer patients who underwent two consecutive mobilizing procedures as part of a tandem transplant protocol. For the first cycle, 10 microg/kg/day granulocyte colony-stimulating factor (G-CSF) was given and apheresis commenced on day 4 and continued for < or =5 days (median 3 days). One week after the first cycle of apheresis, 4000 mg/m2 cyclophosphamide, 400 mg/m2 etoposide, and 10 microg/kg G-CSF were administered for < or =16 days (cycle 2). Apheresis was initiated when the white blood cell (WBC) count exceeded 5000 cells/microL and continued for < or =5 days (median 3 days). Mean values of peripheral blood WBC (31,700+/-3200 vs. 30,700+/-3300/microL) were not significantly different between cycles 1 and 2. Mean number of mononuclear cells (MNC) collected per day was slightly greater with G-CSF mobilization than with the combination of chemotherapy and G-CSF (2.5+/-0.21x10(8) vs. 1.8+/-0.19x10(8) cells/kg). Mean daily CD34+ cell yield, however, was nearly six times higher (12.9+/-4.4 vs. 2.2+/-0.5x10(6)/kg; p = 0.01) with chemotherapy plus G-CSF. With G-CSF alone, 13% of aphereses reached the target dose of 5x10(6) CD34+ cells/kg in one collection vs. 57% with chemotherapy plus G-CSF. Transfusions of red blood cells or platelets were necessary in 18 of 24 patients in cycle 2. Three patients were hospitalized with fever for a median of 3 days after cycle 2. No patients received transfusions or required hospitalization during mobilization with G-CSF alone. Resource utilization (cost of drugs, aphereses, cryopreservation, transfusions, hospitalization) was calculated comparing the median number of collections to obtain a target CD34+ cell dose of 5x10(6) cells/kg: four using G-CSF vs. one using the combination in this data set. Resources for G-CSF mobilization cost $7326 vs. $8693 for the combination, even though more apheresis procedures were performed using G-CSF mobilization. The cost of chemotherapy administration, more doses of G-CSF, transfusions, and hospitalizations caused cyclophosphamide, etoposide, and G-CSF to be more expensive than G-CSF alone. A less toxic and less expensive treatment than cyclophosphamide, etoposide, and G-CSF is needed to be more cost-effective than G-CSF alone for peripheral blood progenitor cell mobilization.

Alterations in predicted growth rates of pediatric kidneys treated with extracorporeal shockwave lithotripsy.

The long-term effects of extracorporeal shockwave lithotripsy (SWL) on the kidneys of children treated for renal calculi are unclear. In order to determine if SWL has any negative effects on renal growth rates, we reviewed long-term (mean 9-year) follow-up data on 29 pediatric patients treated between 1984 and 1988 with an unmodified Dornier HM3 lithotripter. Changes in renal length, serum creatinine, and blood pressure were analyzed. Predicted renal growth was calculated using a formula for age-adjusted renal length. Treated kidneys were stratified into normal and abnormal groups based on a history of renal surgery, evidence of recurrent infection, and obvious anatomic abnormalities. Fifty-six upper urinary tract calculi were treated in 34 renal units. Twenty-two renal units (68%) were rendered stone free, and 65% of the patients continue to be stone free. At follow-up, one patient was classified as having new-onset hypertension, and the mean serum creatinine was 0.93 +/- 0.08 mg/dL. Both at treatment and at follow-up, no significant differences were found in the sizes of the treated and untreated kidneys. However, at treatment, the abnormal group of kidneys seemed to be smaller than expected (mean Z -1.30 +/- 1.10), whereas the group of normal kidneys was very close (mean Z 0.18 +/- 0.54) to the predicted length. At follow-up, the deviations between actual and predicted renal length were significantly more negative. Treated kidneys were an additional 1.26 +/- 0.49 SD units below their expected length (p = 0.02). Untreated kidneys were further below normal as well but possibly to a lesser degree (-0.82 +/- 0.36; p <0.04). Although there was a trend for the abnormal group to have smaller kidneys than the normal group, both groups showed the same trend toward an age-adjusted reduction in renal growth at follow-up. The alterations in renal growth patterns observed in this population are unsettling and could be secondary to either treatment effect (SWL) or, more likely, to some underlying pathology intrinsic to pediatric kidneys with urolithiasis. Until further data are available, SWL in the pediatric population should be applied with caution and at the lowest dosage sufficient to achieve stone comminution.

Factors associated with nursing interventions to reduce incontinence in hospitalized older adults.

INTRODUCTION: The problem of incontinence in hospitalized elderly patients is rarely documented, and there is little research that determines why nurses choose to help or not help with this problem. Are hospital-based acute care nurses' attitudes and beliefs about incontinence associated with the perceived opportunity to assist the patient with the problem? What do hospital nurses know about causes and interventions relative to incontinence? METHODS: Two vignettes, one describing a patient with stress incontinence and one describing a patient with urge incontinence, were created. Questions measuring variables of a help-giving model were developed, and nurses were asked to mark on a Likert-type scale when answering each question. RESULTS: One hundred-fifty respondents returned completed questionnaires along with three nurse experts. Many hospital nurses believed incontinence was temporary and part of being old. As a group they had a more positive attitude toward intervening for urge incontinence and believed the physician and their nurse manager expected them to assist the patient with urge incontinence. Respondents tended to believe the patient was least likely to expect help. Respondents were evenly divided about opportunity to provide assistance for stress or urge incontinence. Less than half of the nurses correctly listed causes and interventions for stress or urge incontinence. CONCLUSIONS: Other clinical problems perceived as more pressing and lack of knowledge concerning appropriate helping measures affect nurses' perceptions of opportunity to intervene when elderly hospital patients are incontinent. Assessment and intervention are essential to quality nursing care. Undergraduate nursing education and ongoing staff education about incontinence are crucial if assessments and interventions are to be correct. Patients, as health care consumers, have to be more educated about incontinence and choose to have the problem addressed during hospitalization. The Agency for Health Care Policy and Research Clinical Practice Guidelines is a major recommended reference.

The use of a break-even analysis: financial analysis of a fast-track program.

OBJECTIVE: To calculate the financial break-even point and illustrate how changes in third-party reimbursement and eligibility could affect a program's fiscal standing. METHODS: Demographic, clinical, and financial data were collected retrospectively for 446 patients treated in a fast-track program during June 1993. The fast-track program is located within the confines of the emergency medicine and trauma center at a 1,050-bed tertiary care Midwestern teaching hospital and provides urgent treatment to minimally ill patients. A financial break-even analysis was performed to determine the point where the program generated enough revenue to cover its total variable and fixed costs, both direct and indirect. RESULTS: Given the relatively low average collection rate (62%) and high percentage of uninsured patients (31%), the analysis showed that the program's revenues covered its direct costs but not all of the indirect costs. CONCLUSIONS: Examining collection rates or payer class mix without examining both costs and revenues may lead to an erroneous conclusion about a program's fiscal viability. Sensitivity analysis also shows that relatively small changes in third-party coverage or eligibility (income) requirements can have a large impact on the program's financial solvency and break-even volumes.

Hospital falls: development of a predictive model for clinical practice.

A retrospective case-control study related to falls was conducted at an 1,120-bed acute care tertiary hospital. The case (fall) sample consisted of 102 falls and 236 control (nonfall) charts during a 1-month period. An instrument developed by Hendrich (1988) was modified for use in the study. Demographic data and risk factors were recorded. Descriptive statistics included risk factor percentages for each sample and the corresponding univariate relative risks. Logistic regression was used to develop a multivariate risk factor model with seven risk factors. The significant risk factors were recent history of falls, depression, altered elimination patterns, dizziness or vertigo, primary cancer diagnosis, confusion, and altered mobility. The adjusted relative risks were converted to risk points to be used to assess a patient's level of fall risk. Within the data set, a sensitivity of 77% (79 of 102) and specificity of 72% (169 of 236) were calculated. The model was cross-validated in a 1987 data set with a sensitivity of 83% (59 of 71) and specificity of 66% (106 of 161).

Community hospital and medical school cooperation in continuing medical education.

PURPOSE: To determine the extent and trends of cooperation in continuing medical education (CME) between community teaching hospitals and medical schools in the United States. METHOD: A questionnaire was sent in September 1992 to the directors of CME at 276 teaching hospital members of the Association for Hospital Medical Education (AHME). The survey was designed to answer two questions: (1) What is the extent of cooperation between hospital CME providers and medical schools? (2) In the next three years will community hospitals seek competitive or collaborative relationships in CME with medical schools? RESULTS: By late April 1993, 216 (78%) of the questionnaires had been returned. Of these, 177 (64% of the sample) were analyzed. Of the responding hospitals, 91 (52%) cooperated with 92 medical schools in CME; 75 (45%) of the hospitals planned to increase cooperation. Only ten (11%) of the hospitals described their current CME relationship with a medical school as "competitive in most areas"; 23 (14%) expected to increase competition in the next three years. Forty-one (24%) of the respondents were part of a community hospital CME consortium; only 20 (16%) of the other institutions expected to participate in a consortium in the next three years. Hospital size and membership in the Association of American Medical Colleges' Council of Teaching Hospitals were generally correlated with current and future competition in CME with a medical school and likely participation in a community CME consortium. CONCLUSION: The majority of teaching hospital members of the AHME perceived that they would have cooperative relationships in CME with affiliated medical schools in the three years following the survey. These collaborative relationships should provide an important basis for the further planning and development of medical education consortia.

Does anybody really know what time it is? Does anybody really care?

STUDY OBJECTIVE: To evaluate the difference among time sources in an emergency medical system. DESIGN: Prospective; comparison to a criterion standard. SETTING: Five emergency departments and three emergency medical services systems in Indianapolis, Indiana. INTERVENTIONS: Coordinated Universal Time (UTC), generated by the atomic clock in Boulder, Colorado, and broadcast by the US Commerce Department's National Institute of Standards and Technology, was used as the time standard. The investigators, on a single day, made unannounced visits to the five EDs and the ambulances and fire stations in the three emergency medical services systems. The times displayed on all time sources at each location were recorded. The accuracy to the second of each time source compared to UTC was calculated. RESULTS: Three time sources were excluded (two defibrillator clocks and one ED wall clock that varied more than three hours from UTC). Of the 152 time sources, 72 had analog displays, 74 digital, three both, and three other. The average absolute difference from UTC was 1 minute 45 seconds (SEM, 9 seconds) with a range of 12 minutes 34 seconds slow to 7 minutes 7 seconds fast. Thus, two timepieces could have varied by as much as 19 minutes 41 seconds. Compared to UTC, 47 timepieces (31%) were slow, 100 (66%) were fast, and five (3%) were accurate to the second. Fifty-five percent of the time sources varied one minute or more from UTC. CONCLUSION: Time sources in this health care system varied considerably. Time recording in medicine could be made more precise by synchronizing medical clocks to UTC, using computers to automatically "time stamp" data entries and using only digital time sources with second displays.

Management of lower pole nephrolithiasis: a critical analysis.

The results of extracorporeal shock wave lithotripsy (ESWL*) and percutaneous nephrostolithotomy for the treatment of lower pole nephrolithiasis were examined in 32 consecutive patients undergoing percutaneous nephrostolithotomy at the Methodist Hospital of Indiana and through meta-analysis of publications providing adequate stratification of treatment results. Of 101 cases managed with percutaneous nephrostolithotomy 91 (90%) were stone-free, a result significantly better than that achieved with ESWL (1,733 of 2,927 stone-free, 59%). Stone-free rates with percutaneous nephrostolithotomy were independent of stone burden, whereas stone-free rates with ESWL were inversely correlated to the stone burden treated. The morbidity of patients undergoing percutaneous nephrostolithotomy at our hospital was minimal, with a mean hospital stay of 4.7 +/- 2.8 days. No blood transfusions were required. All patients became stone-free. The percentage of urolithiasis patients with lower pole calculi is increasing. Because of the significantly greater efficacy of percutaneous nephrostolithotomy for lower pole calculi, particularly stones larger than 10 mm. in diameter, further consideration should be given to an initial approach with percutaneous nephrostolithotomy.

Accuracy of bar codes versus handwriting for recording trauma resuscitation events.

STUDY OBJECTIVES: To compare the accuracy of computerized bar code data entry with conventional handwritten data entry during videotaped trauma resuscitations. SETTING: Laboratory; video simulation. TYPE OF PARTICIPANTS: Twenty-four emergency nurses. DESIGN: The nurses viewed videotapes of four different major trauma resuscitations during a single session. Each nurse recorded resuscitation events by handwritten entry in two cases and by bar code entry in another two. A computerized bar code system was designed specifically for contemporaneous charting of rapidly occurring events during trauma resuscitations. The handwritten and bar-coded records then were compared with a master list of events, and the number of entry errors were counted. Errors were defined as "omissions" (failing to record an event), "commissions" (recording an event that did not occur), or "inaccuracies" (errors in recording details of an event). ANALYSIS: Differences in the number of entry errors between the two recording methods were compared using unpaired t-tests. Differences in the number of errors after adjusting for the different nurses, different case being viewed, and order of viewing were analyzed using balanced analysis of variance techniques. P < .05 was considered significant. MAIN RESULTS: The mean +/- SEM number of total errors per record for bar codes was 2.63 +/- 0.24 compared with 4.48 +/- 0.30 for handwriting (P < .0001). The mean number of omissions per record for bar codes was 2.25 +/- 0.21 compared with 3.65 +/- 0.27 for handwriting (P = .0001). The mean number of inaccuracies per record for bar codes was 0.38 +/- 0.10 compared with 0.83 +/- 0.12 for handwriting (P = .0038). There were no commission-type errors. CONCLUSION: Computerized bar code data entry of trauma resuscitation events had significantly fewer entry errors than handwritten data entry in a laboratory setting. Potential advantages of bar code data entry include keyless data entry, automatic time-stamping, standardization of documentation, legibility of the medical record, and "point-of-care" data capture.

Changes in serum catecholamine levels in patients who are brain dead.

Prospective blood samplings from 15 patients admitted with a Glasgow Coma Score of less than 7 were obtained to observe and compare epinephrine, norepinephrine, and dopamine serum levels in patients with brain injury before, after, and in the absence of brain death. Nine of the patients developed or were admitted after brain death. Wide variations in catecholamine blood levels over time were documented, and subgroup analysis precluded useful statistical comparison or inference of the data. The data are presented therefore as descriptive observations only. No apparent differences were noted between similarly injured patients in whom brain death did not develop and patients before brain death or between patients with penetrating versus nonpenetrating brain injury. Brain death was preceded by hypertension and corresponding elevations in serum catecholamine levels in one patient with complete data. Catecholamine levels appeared to fall after brain death in most patients. Only minimal changes in myocardial histology were present in three donor hearts, and the two transplanted hearts functioned satisfactorily. Serum catecholamine measurement or monitoring does not provide a precise method of determining potential injury to the donor heart before or after brain death. Other experimental data and clinical observation indicate that some hearts may be injured in the donor during the evolution of brain death. Pharmacologic intervention may prevent such injury in experimental animals but must be used before brain death is induced. Such interventions should be studied in selected human donors before brain death to determine whether cardiac function is improved in the donor or recipient.

The learning curve and the cost of heart transplantation.

The effect of learning on hospital outcomes such as mortality or adverse events (the so-called "practice makes perfect" hypothesis) has been studied by numerous investigators. The effect of learning on hospital cost, however, has received much less attention. This article reports the results of a multiple regression model demonstrating a nonlinear, decreasing trend in operative and postoperative hospital costs over time in a consecutive series of 71 heart transplant patients, all treated in the same institution. The cost trend is shown to persist even after controlling for various preoperative demographic and clinical risk factors and the specific experience of individual surgeons. Using a reference case, the model predicts a cost of $81,297 for the first heart transplant procedure performed at the hospital. If this same patient had been the tenth case rather than the first, with the hospital having benefited from the experience gained in nine previous cases, the model predicts the cost would now be only $48,431, or approximately 60 percent of the cost of the first case. Had this patient been the twenty-fifth case, the predicted cost would be $35,352 (43 percent of the original cost), and had this been the fiftieth case, the cost would be $25,458 (31 percent of the original cost). The longitudinal study design used in this analysis greatly reduces the likelihood that the observed cost reduction is due to economies of scale rather than learning. The results have implications for a policy of regionalization as a tactic for containing hospital cost. Whereas others have pointed to a volume-cost relationship as an argument for the regionalization of expensive and complex hospital procedures, the present data isolate a learning-cost relationship as a separate argument for regionalization.

Staghorn calculi: analysis of treatment results between initial percutaneous nephrostolithotomy and extracorporeal shock wave lithotripsy monotherapy with reference to surface area.

Treatment recommendations and results reported for the management of staghorn calculi are highly variable. In an attempt to provide a more objective means to compare treatment results for staghorn renal calculi, stone burden as measured by stone surface area was used. Stone surface area was determined by computer analysis. A total of 380 cases of staghorn calculi treated at the same institution was evaluated. Treatment consisted of initial percutaneous nephrostolithotomy with or without extracorporeal shock wave lithotripsy (ESWL*) in 298 cases and ESWL monotherapy in 82. When considered as a group, the overall stone-free rate for initial percutaneous nephrostolithotomy (mean surface area 1,378.3 mm.2) was 84.2% compared to 51.2% (p less than 0.0001) for ESWL monotherapy (mean surface area 693.4 mm.2). For staghorn calculi smaller than 500 mm.2 a stone-free rate of 94.4% was achieved in the percutaneous nephrostolithotomy with or without ESWL group compared to 63.2% for ESWL monotherapy (p = 0.0214). For calculi of 501 to 1,000 mm.2 the stone-free rates were 86% and 45.7%, respectively (p less than 0.0001). When stone surface area exceeded 1,000 mm.2 the stone-free rate for percutaneous nephrostolithotomy with or without ESWL was 82.4% but it was only 22.2% for ESWL monotherapy (p = 0.0002). Overall, when adjusted for stone surface area the odds of being stone-free were more than 8 times higher for initial percutaneous nephrostolithotomy versus ESWL monotherapy (odds ratio = 8.36, p less than 0.0001). While percutaneous nephrostolithotomy with or without ESWL appears to be the procedure of choice for most staghorn stones, ESWL monotherapy may have a role for some stones smaller than 500 mm.2. In 12 such cases associated with a nondilated renal collecting system (mean surface area 380.5 mm.2) a stone-free rate of 91.7% was achieved. The number of procedures required to complete therapy was higher in the initial percutaneous nephrostolithotomy group (2.8 versus 2.1, p less than 0.0001). Although complications were more common in the ESWL monotherapy group (manifested as obstruction in 30.5%), bleeding requiring blood transfusion was more frequent in the initial percutaneous nephrostolithotomy group (9.4%).

An analysis of reimbursement for outpatient medical care in an urban hospital emergency department.

The investigators examined the demographic and clinical factors associated with the collection experience in a series of 786 patients who were treated in an urban hospital emergency department (ED) but not admitted to the hospital. They found that 57% of the total net charge of $150,489 had been paid within 180 days. This rate can be compared with an average inpatient collection rate of 85% at 180 days. Seven factors were found to account for the collection rate variation, making up 38.4% of the total variation. Age, gender, primary diagnosis, season of visit, time of arrival, and residence were not found to be main contributors. Insufficient collection rates may be an indication that EDs increasingly are becoming a financial risk to hospitals. The hospital's collection experience will become more important as an indicator of financial risk if the costs of operating EDs continue to escalate and collection rates do not improve. Both the costs of providing a service and the amount of the charge actually collected are valid concerns to those operating EDs.

Cardiac pathology in 470 consecutive forensic autopsies.

Cardiovascular disease continues to be the single most common generic cause of sudden and unexpected deaths. Atherosclerotic coronary heart disease and acute myocardial infarction are the most prevalent forms of fatal cardiac disease observed at autopsy. Other cardiac lesions are frequently listed as causes of death, but the prevalence of such lesions as incidental findings in the general population is unknown. In this study, 470 consecutive forensic autopsies were evaluated for minor and major anomalies. The most frequently observed major congenital finding was floppy mitral valve (5%). Tunneled coronary arteries, considered minor congenital findings, were seen in 29%. Atherosclerotic coronary heart disease was the most common major acquired finding, observed in 16% of cases. Of the 470 hearts, only 8% were considered normal.

Orthopedic research: an overview of data entry, database management, and statistical analysis.

An orthopedic practitioner can facilitate clinical research and analyze quality assurance data with a minor investment in a personal computer, an optical scanner, and two software packages, namely a database manager and a statistics program. One of the most time-consuming stages in the research process includes entering patient chart data, editing and manipulating the data (database management), and analyzing the data (statistical analysis). This can be automated to a large extent with the above mentioned equipment. This article focuses on the steps involved in organizing an orthopedic office for research. The steps include choosing a method of data entry, choosing and implementing a database package, and choosing and implementing a statistics package. This discussion is followed by a practical review of basic statistics applicable to orthopedic research. Several simple and advanced tests are described and examples are given for each.

Financial analysis of an inner-city helicopter service: charges versus collections for patients transferred from another acute care facility.

Emergency helicopter services provided by trauma centers are now being perceived as contributing to the financial burden of the hospital because of recent changes in trauma reimbursement under the Medicare Prospective Payment System (PPS) and because of the general perception that collection rates are lower among trauma patients. The use of helicopters to transfer patients from one acute care facility to another may also be concentrating the patients with low collection rates at the receiving hospital. We examined retrospectively the demographic and clinical factors associated with the collection experience in a series of 288 trauma patients transferred by helicopter from another acute care facility to an inner-city hospital. Factors affecting payment at 180 and 360 days included patient age, insurance class, discharge status, and size of the hospital charge. As long as reimbursement continues to be cost-based for nonMedicare patients, collection rates remain an important consideration in determining the financial viability of using helicopters to transfer patients.

Cost analysis of heart transplantation from the day of operation to the day of discharge.

Hospital costs from the day of transplantation to the day of discharge were examined in a consecutive series of 53 patients who underwent orthotopic heart transplantation between October 1982 and February 1987. An accounting cost methodology was used to convert billable charges, to costs for 29 separate hospital cost centers. Total cost per case has shown a statistically significant decrease of over $30,000 with no indication of a change in patient selection or a decrease in 3-month survival. Most of the cost reductions occurred in five cost centers: operating room, blood and intravenous therapy, medical supplies, heart station, and routine services, as evidenced by decreases in wages and supplies. The results support the premise that new technologies can become more cost-efficient over time and suggest that as the medical team becomes more proficient and experienced, cost reductions can become a reality.

Financial analysis of an inner-city helicopter service: charges versus collections.

Trauma centers are now being perceived as financial burdens because of recent changes in trauma reimbursement for the Medicare Prospective Payment System population and the perception that collection rates are lower among trauma patients. We examined the demographic and clinical factors associated with the collection experience in a series of 114 trauma patients transferred by helicopter from the accident site to an inner-city trauma center. Factors affecting payment at 30, 60, 90, and 180 days included patient age, insurance class, and discharge status. While not as high as the collection rate for the facility as a whole, we found an average 71.2% collection rate for trauma patients at 180 days. As long as trauma reimbursement continues to be cost based for nonMedicare patients, collection rates remain an important consideration in determining the financial viability of trauma centers.

Comparison of the effects of pindolol and atenolol on hemodynamic function in systemic hypertension.

A randomized double-blind study was performed on a group of mild hypertensive patients (WHO class I) to compare the hemodynamic effects of pindolol and atenolol. Blood pressure (BP) was monitored with a mercury gauge sphygmomanometer, while cardiac function and peripheral arterial flows were measured by the noninvasive technique of bioelectric impedance. After a 2-week washout period, patients with a diastolic BP greater than 95 mm Hg but less than 114 mm Hg were randomized into the pindolol (29 patients) or atenolol (28) treatment groups. Patients were treated with 1 of the 2 drugs in an incremental fashion for 12 weeks. Cardiovascular function was measured after the washout period and at the end of the 12-week treatment period. Baseline hemodynamics were similar in both groups. The 2 drugs were equally effective in lowering both systolic and diastolic BP. Hemodynamically, pindolol lowered BP by decreasing total peripheral resistance (-406 +/- 145 dynes.s.cm-5) while atenolol decreased cardiac index (-0.2 +/- 0.1 liters/min/m2) associated with a decrease in heart rate (-12 +/- 2 beats/min). Regarding peripheral vascular beds, pindolol lowered arm vascular resistance (-198 +/- 72 mm Hg/liter/min) and leg vascular resistance (-73 +/- 25 mm Hg/liter/min), especially when subjects who did not respond to pindolol were excluded from the analysis. Both arm (5.5 +/- 5.4% increase above baseline) and leg (1.2 +/- 4.4% increase above baseline) arterial flow indexes were maintained with pindolol. Conversely, atenolol decreased the arm arterial flow index (-9,8 +/- 5.6% decrease below baseline), but not significantly and with no change in resistance (+54 +/- 62 mm Hg/liter/min).(ABSTRACT TRUNCATED AT 250 WORDS)