Predicting reamputation risk in patients undergoing lower extremity amputation due to the complications of peripheral artery disease and/or diabetes

Background

Patients undergoing amputation of the lower extremity for the complications of peripheral artery disease and/or diabetes are at risk of treatment failure and the need for reamputation at a higher level. The aim of this study was to develop a patient-specific reamputation risk prediction model.

Methods

Patients with incident unilateral transmetatarsal, transtibial or transfemoral amputation between 2004 and 2014 secondary to diabetes and/or peripheral artery disease, and who survived 12 months after amputation, were identified using Veterans Health Administration databases. Procedure codes and natural language processing were used to define subsequent ipsilateral reamputation at the same or higher level. Stepdown logistic regression was used to develop the prediction model. It was then evaluated for calibration and discrimination by evaluating the goodness of fit, area under the receiver operating characteristic curve (AUC) and discrimination slope.

Results

Some 5260 patients were identified, of whom 1283 (24·4 per cent) underwent ipsilateral reamputation in the 12 months after initial amputation. Crude reamputation risks were 40·3, 25·9 and 9·7 per cent in the transmetatarsal, transtibial and transfemoral groups respectively. The final prediction model included 11 predictors (amputation level, sex, smoking, alcohol, rest pain, use of outpatient anticoagulants, diabetes, chronic obstructive pulmonary disease, white blood cell count, kidney failure and previous revascularization), along with four interaction terms. Evaluation of the prediction characteristics indicated good model calibration with goodness-of-fit testing, good discrimination (AUC 0·72) and a discrimination slope of 11·2 per cent.

Conclusion

A prediction model was developed to calculate individual risk of primary healing failure and the need for reamputation surgery at each amputation level. This model may assist clinical decision-making regarding amputation-level selection.

Mortality prediction following non-traumatic amputation of the lower extremity

BACKGROUND

Patients who undergo lower extremity amputation secondary to the complications of diabetes or peripheral artery disease have poor long-term survival. Providing patients and surgeons with individual-patient, rather than population, survival estimates provides them with important information to make individualized treatment decisions.

METHODS

Patients with peripheral artery disease and/or diabetes undergoing their first unilateral transmetatarsal, transtibial or transfemoral amputation were identified in the Veterans Affairs Surgical Quality Improvement Program (VASQIP) database. Stepdown logistic regression was used to develop a 1-year mortality risk prediction model from a list of 33 candidate predictors using data from three of five Department of Veterans Affairs national geographical regions. External geographical validation was performed using data from the remaining two regions. Calibration and discrimination were assessed in the development and validation samples.

RESULTS

The development sample included 5028 patients and the validation sample 2140. The final mortality prediction model (AMPREDICT-Mortality) included amputation level, age, BMI, race, functional status, congestive heart failure, dialysis, blood urea nitrogen level, and white blood cell and platelet counts. The model fit in the validation sample was good. The area under the receiver operating characteristic (ROC) curve for the validation sample was 0·76 and Cox calibration regression indicated excellent calibration (slope 0·96, 95 per cent c.i. 0·85 to 1·06; intercept 0·02, 95 per cent c.i. -0·12 to 0·17). Given the external validation characteristics, the development and validation samples were combined, giving a total sample of 7168.

CONCLUSION

The AMPREDICT-Mortality prediction model is a validated parsimonious model that can be used to inform the 1-year mortality risk following non-traumatic lower extremity amputation of patients with peripheral artery disease or diabetes.

Randomized Trial of Motivational Interviewing to Prevent Early Childhood Caries in American Indian Children

Introduction

In a randomized controlled trial, the effectiveness of motivational interviewing (MI) combined with enhanced community services (MI + ECS) was compared with ECS alone for reducing dental caries in American Indian children on the Pine Ridge Reservation. The intervention was developed and delivered with extensive tribal collaboration.

Methods

A total 579 mother-newborn dyads were enrolled and randomized to the MI + ECS and ECS groups. They were followed for 36 mo. Four MI sessions were provided, the first shortly after childbirth and then 6, 12, and 18 mo later. Both groups were exposed to ECS, which included public service announcements through billboards and tribal radio, as well as broad distribution of brochures on behavioral risk factors for early childhood caries (ECC), toothbrushes, and toothpaste. MI impact was measured as decayed, missing, and filled tooth surfaces (dmfs). Secondary outcomes included decayed surfaces, caries prevalence, and maternal oral health knowledge and behaviors. Modified intention-to-treat analyses were conducted. Eighty-eight percent of mothers completed at least 3 of 4 MI sessions offered.

Results

After 3 y, dmfs was not significantly different for the 2 groups (MI + ECS = 10, ECS = 10.38, P = 0.68). In both groups, prevalence of caries experience was 7% to 9% after 1 y, 35% to 36% at 2 y, and 55% to 56% at 3 y. Mean knowledge scores increased by 5.0, 5.3, and 5.9 percentage points at years 1, 2, and 3 in the MI + ECS group and by 1.9, 3.3, and 5.0 percentage points in the ECS group (P = 0.03), respectively. Mean maternal oral health behavior scores were not statistically significantly different between the treatment arms.

Conclusion

In summary, the MI intervention appeared to improve maternal knowledge but had no effect on oral health behaviors or on the progression of ECC (ClinicalTrials.gov NCT01116726).

Knowledge Transfer Statement

The findings of this study suggest that motivational interviewing focusing on parental behaviors may not be as effective as previously hoped for slowing the development of childhood caries in some high-risk groups. Furthermore, social factors may be even more salient determinants of oral health than what we previously supposed, perhaps interfering with the capacity to benefit from behavioral strategies that have been useful elsewhere. The improvement of children's oral health in high-risk populations characterized by poverty and multiple related life stresses may require more holistic approaches that address these formidable barriers.

A Cluster-Randomized, Community-Based, Tribally Delivered Oral Health Promotion Trial in Navajo Head Start Children

The authors tested the effectiveness of a community-based, tribally delivered oral health promotion (OHP) intervention (INT) at reducing caries increment in Navajo children attending Head Start. In a 3-y cluster-randomized trial, we developed an OHP INT with Navajo input that was delivered by trained Navajo lay health workers to children attending 52 Navajo Head Start classrooms (26 INT, 26 usual care [UC]). The INT was designed as a highly personalized set of oral health-focused interactions (5 for children and 4 for parents), along with 4 fluoride varnish applications delivered in Head Start during academic years of 2011 to 2012 and 2012 to 2013. The authors evaluated INT impact on decayed, missing, and filled tooth surfaces (dmfs) increment compared with UC. Other outcomes included caries prevalence and caregiver oral health-related knowledge and behaviors. Modified intention-to-treat and per-protocol analyses were conducted. The authors enrolled 1,016 caregiver-child dyads. Baseline mean dmfs/caries prevalence equaled 19.9/86.5% for the INT group and 22.8/90.1% for the UC group, respectively. INT adherence was 53% (i.e., ≥3 child OHP events, ≥1 caregiver OHP events, and ≥3 fluoride varnish). After 3 y, dmfs increased in both groups (+12.9 INT vs. +10.8 UC; P = 0.216), as did caries prevalence (86.5% to 96.6% INT vs. 90.1% to 98.2% UC; P = 0.808) in a modified intention-to-treat analysis of 897 caregiver-child dyads receiving 1 y of INT. Caregiver oral health knowledge scores improved in both groups (75.1% to 81.2% INT vs. 73.6% to 79.5% UC; P = 0.369). Caregiver oral health behavior scores improved more rapidly in the INT group versus the UC group (P = 0.006). The dmfs increment was smaller among adherent INT children (+8.9) than among UC children (+10.8; P = 0.028) in a per-protocol analysis. In conclusion, the severity of dental disease in Navajo Head Start children is extreme and difficult to improve. The authors argue that successful approaches to prevention may require even more highly personalized approaches shaped by cultural perspectives and attentive to the social determinants of oral health (ClinicalTrials.gov NCT01116739).

Association of parental health literacy with oral health of Navajo Nation preschoolers

Health literacy is 'the capacity to obtain, process and understand basic health information and services needed to make appropriate health decisions'. Although numerous studies show a link between health literacy and clinical outcomes, little research has examined the association of health literacy with oral health. No large-scale studies have assessed these relationships among American Indians, a population at risk for limited health literacy and oral health problems. This analysis was conducted as part of a clinical trial aimed at reducing dental decay among preschoolers in the Navajo Nation Head Start program. Using baseline data for 1016 parent-child dyads, we examined the association of parental health literacy with parents' oral health knowledge, attitudes, and behavior, as well as indicators of parental and pediatric oral health. More limited health literacy was associated with lower levels of oral health knowledge, more negative oral health attitudes, and lower levels of adherence to recommended oral health behavior. Parents with more limited health literacy also had significantly worse oral health status (OHS) and reported their children to have significantly worse oral health-related quality of life. These results highlight the importance of oral health promotion interventions that are sensitive to the needs of participants with limited health literacy.

Changes in health-related quality of life following coronary artery bypass graft surgery

BACKGROUND

There are limited data to help clinicians identify patients likely to have an improvement in quality of life following CABG surgery. We evaluated the relationship between preoperative health status and changes in quality of life following CABG surgery.

METHODS

We evaluated 1,744 patients enrolled in the VA Cooperative Processes, Structures, and Outcomes in Cardiac Surgery study who completed preoperative and 6-month postoperative Short Form-36 (SF-36) surveys. The primary outcome was change in the Mental Component Summary (MCS) and Physical Component Summary (PCS) scores from the SF-36.

RESULTS

On average, physical and mental health status improved following the operation. Preoperative health status was the major determinant of change in quality of life following surgery, independent of anginal burden and other clinical characteristics. Patients with MCS scores less than 44 or PCS scores less than 38 were most likely to have an improvement in quality of life. Patients with higher preoperative scores were unlikely to have an improvement in quality of life.

CONCLUSIONS

Patients with preoperative health status deficits are likely to have an improvement in their quality of life following CABG surgery. Alternatively, patients with relatively good preoperative health status are unlikely to have a quality of life benefit from surgery and the operation should primarily be performed to improve survival.

Development and validation of a multifactorial risk index for predicting postoperative pneumonia after major noncardiac surgery

BACKGROUND

Pneumonia is a common postoperative complication associated with substantial morbidity and mortality.

OBJECTIVE

To develop and validate a preoperative risk index for predicting postoperative pneumonia.

DESIGN

Prospective cohort study with outcome assessment based on chart review.

SETTING

100 Veterans Affairs Medical Centers performing major surgery.

PATIENTS

The risk index was developed by using data on 160 805 patients undergoing major noncardiac surgery between 1 September 1997 and 31 August 1999 and was validated by using data on 155 266 patients undergoing surgery between 1 September 1995 and 31 August 1997. Patients with preoperative pneumonia, ventilator dependence, and pneumonia that developed after postoperative respiratory failure were excluded.

MEASUREMENTS

Postoperative pneumonia was defined by using the Centers for Disease Control and Prevention definition of nosocomial pneumonia.

RESULTS

A total of 2466 patients (1.5%) developed pneumonia, and the 30-day postoperative mortality rate was 21%. A postoperative pneumonia risk index was developed that included type of surgery (abdominal aortic aneurysm repair, thoracic, upper abdominal, neck, vascular, and neurosurgery), age, functional status, weight loss, chronic obstructive pulmonary disease, general anesthesia, impaired sensorium, cerebral vascular accident, blood urea nitrogen level, transfusion, emergency surgery, long-term steroid use, smoking, and alcohol use. Patients were divided into five risk classes by using risk index scores. Pneumonia rates were 0.2% among those with 0 to 15 risk points, 1.2% for those with 16 to 25 risk points, 4.0% for those with 26 to 40 risk points, 9.4% for those with 41 to 55 risk points, and 15.3% for those with more than 55 risk points. The C-statistic was 0.805 for the development cohort and 0.817 for the validation cohort.

CONCLUSIONS

The postoperative pneumonia risk index identifies patients at risk for postoperative pneumonia and may be useful in guiding perioperative respiratory care.

Preoperative risk factors for 30-day mortality after elective surgery for vascular disease in Department of Veterans Affairs hospitals: is race important?

PURPOSE

Racial variation in health care outcomes is an important topic. Risk-adjustment models have not been developed for elective abdominal aortic aneurysm repair (AAA), lower extremity bypass revascularization (LEB), or lower extremity amputation (AMP). Earlier studies examining racial variation in mortality and morbidity from AAA, LEB, or AMP were limited to administrative data. This study determined risk factors for mortality after surgery for vascular disease and determined whether race is an important risk factor.

METHODS

Data in this prospective observational study were obtained from the Department of Veterans Affairs (VA) National Surgical Quality Improvement Program. Detailed demographic and clinical data were collected prospectively from patients' medical records by trained nurse reviewers. Eligible patients were those 18 years and older who underwent elective AAA, LEB, or AMP at one of 44 VA medical centers performing both vascular and cardiac surgery (phase I; October 1991 to December 1993) and at one of these 44 or 79 additional VA medical centers performing vascular but not cardiac surgery (phase II; January 1994 to August 1995). The independent association of several preoperative factors with the 30-day postoperative mortality rate was examined with stepwise logistic regression analysis for AAA, LEB, and AMP. Models were developed in the combined 44 VA medical centers and validated in the 79 VA medical centers. The independent association of race with the 30-day postoperative mortality rate was examined after controlling for important preoperative risk factors for each operation.

RESULTS

More than 10,000 surgical operations were examined, and 5, 3, and 10 independent preoperative predictors of 30-day mortality rate were identified for AAA, LEB, and AMP, respectively. The observed mortality rate for patients undergoing AAA was higher (7.2% vs 3.2%; P =.02) in African American patients than in white patients in the 44 VA medical centers, although the differences were not significant in LEB and AMP or at the additional 79 hospitals. After important preoperative risk factors were controlled, there was no difference in 30-day mortality rates between African American patients and white patients.

CONCLUSION

We identified several important preoperative risk factors for 30-day mortality rate in three vascular operations. From the results of this study, race was determined not to be an independent predictor of mortality.

Comparison of surgical outcomes between teaching and nonteaching hospitals in the Department of Veterans Affairs

OBJECTIVE

To determine whether the investment in postgraduate education and training places patients at risk for worse outcomes and higher costs than if medical and surgical care was delivered in nonteaching settings.

SUMMARY BACKGROUND DATA

The Veterans Health Administration (VA) plays a major role in the training of medical students, residents, and fellows.

METHODS

The database of the VA National Surgical Quality Improvement Program was analyzed for all major noncardiac operations performed during fiscal years 1997, 1998, and 1999. Teaching status of a hospital was determined on the basis of a background and structure questionnaire that was independently verified by a research fellow. Stepwise logistic regression was used to construct separate models predictive of 30-day mortality and morbidity for each of seven surgical specialties and eight operations. Based on these models, a severity index for each patient was calculated. Hierarchical logistic regression models were then created to examine the relationship between teaching versus nonteaching hospitals and 30-day postoperative mortality and morbidity, after adjusting for patient severity.

RESULTS

Teaching hospitals performed 81% of the total surgical workload and 90% of the major surgery workload. In most specialties in teaching hospitals, the residents were the primary surgeons in more than 90% of the operations. Compared with nonteaching hospitals, the patient populations in teaching hospitals had a higher prevalence of risk factors, underwent more complex operations, and had longer operation times. Risk-adjusted mortality rates were not different between the teaching and nonteaching hospitals in the specialties and operations studied. The unadjusted complication rate was higher in teaching hospitals in six of seven specialties and four of eight operations. Risk adjustment did not eliminate completely these differences, probably reflecting the relatively poor predictive validity of some of the risk adjustment models for morbidity. Length of stay after major operations was not consistently different between teaching and nonteaching hospitals.

CONCLUSION

Compared with nonteaching hospitals, teaching hospitals in the VA perform the majority of complex and high-risk major procedures, with comparable risk-adjusted 30-day mortality rates. Risk-adjusted 30-day morbidity rates in teaching hospitals are higher in some specialties and operations than in nonteaching hospitals. Although this may reflect the weak predictive validity of some of the risk adjustment models for morbidity, it may also represent suboptimal processes and structures of care that are unique to teaching hospitals. Despite good quality of care in teaching hospitals, as evidenced by the 30-day mortality data, efforts should be made to examine further the structures and processes of surgical care prevailing in these hospitals.

Postoperative and amputation-free survival outcomes after femorodistal bypass grafting surgery: findings from the Department of Veterans Affairs National Surgical Quality Improvement Program

PURPOSE

A noncardiac surgery risk model was used as a means of analyzing variations in postoperative mortality and amputation-free survival for older veterans undergoing femorodistal bypass grafting surgery.

METHODS

A prospective cohort study was undertaken in 105 Veterans Affairs (VA) hospitals at the time of index operation from 1991 to 1995. Each patient was linked to subsequent hospitalizations, major amputation surgery, and survival through 1999. Logistic regression and proportional hazards models were used as a means of developing risk indices on the basis of risk factors from the VA National Surgical Quality Improvement Program. A total of 4288 male veterans 40 years or older underwent artificial, vein, or in situ bypass grafting surgery at the femoral to tibial level. The main outcome measures were 30-day postoperative mortality and amputation-free survival.

RESULTS

Approximately half of all patients had undergone an earlier revascularization or amputation at any level for vascular disease. The 30-day postoperative mortality rate was 2.1% and varied greatly between mortality risk index quartiles (0.6%-5.2%). In a median 44.3 months of follow-up, surviving patients had 17,694 subsequent VA hospitalizations, 1147 patients (26.7%) underwent subsequent major amputation, and 1913 patients (44.6%) died. The overall survival probability was 88% at 1 year and 63% at 5 years; 1- and 5-year (any sided) limb salvage rates were 87% and 74%, respectively, for patients who underwent a femoropopliteal bypass grafting procedure, compared with 77% and 63%, respectively, for patients who underwent a tibial bypass grafting procedure. When amputation and death were combined as end points, amputation-free survival probability rates at 1, 3, and 7.5 years were 74%, 56%, and 29%, respectively. Patients with the best 20% survival risk scores had observed mean survival probability rates 30% higher than patients in the poorest 20% of survival risk.

CONCLUSION

Risk indices derived from the preoperative workup may be of use to clinicians in assessing and communicating risk and prognosis. Risk-adjustment of outcomes is critical for evaluating future disease management initiatives for patients with advanced peripheral arterial disease.

Postoperative and late survival outcomes after major amputation: findings from the Department of Veterans Affairs National Surgical Quality Improvement Program

BACKGROUND

A surgical risk model is used to analyze postoperative mortality and late survival for older veterans who underwent above- or below-knee amputations in 119 Veterans Affairs (VA) hospitals from 1991 to 1995.

METHODS

Preoperative medical conditions and laboratory values abstracted by the VA National Surgical Quality Improvement Program were linked to subsequent hospitalization and survival through 1999. Logistic regression and proportional hazards models were used to develop risk indexes for postoperative mortality and long-term survival.

RESULTS

Thirty-day postoperative mortality was 6.3% for 1909 below-knee and 13.3% for 2152 above-knee amputees. Mortality varied greatly between the lowest-highest risk index quartiles (0.8%-18.4% for below-knee amputation and 2.3%-31.1% for above-knee amputation). Surviving patients had 10,827 subsequent VA hospitalizations during a median 32-month follow-up. Survival probabilities for below- and above-knee amputees were 77% and 59% at 1 year, 57% and 39% at 3 years, and 28% and 20% at 7.5 years. The lowest quartile of survival risk had a 61% five-year survival compared with 14% for the highest-risk quartile.

CONCLUSION

A generic surgical risk model can be of use in stratifying prognosis after major amputation. The heavy burden of hospital use by these patients suggests the need for better disease management for this high-risk, high-cost patient population.

Diagnostic value of iron indices in hemodialysis patients receiving epoetin

BACKGROUND

Iron deficiency remains a common cause of hyporesponsiveness to epoetin in hemodialysis patients. However, considerable controversy exists regarding the best strategies for diagnosis and treatment.

METHODS

As part of a multicenter randomized clinical trial of intravenous versus subcutaneous administration of epoetin, we made monthly determinations of serum iron, total iron binding capacity, percentage transferrin saturation, and serum ferritin. If a patient had serum ferritin <100 ng/mL or the combination of serum ferritin <400 ng/mL and a transferrin saturation <20%, he/she received parenteral iron, given as iron dextran 100 mg at ten consecutive dialysis sessions. We analyzed parenteral iron use during the trial, the effect of its administration on iron indices and epoetin dose, and the ability of the iron indices to predict a reduction in epoetin dose in response to parenteral iron administration.

RESULTS

Eighty-seven percent of the 208 patients required parenteral iron to maintain adequate iron stores at an average dose of 1516 mg over 41.7 weeks, or 36 mg/week. Only two of 180 patients experienced serious reactions to intravenous iron administration. Two thirds of the patients receiving parenteral iron had a decrease in their epoetin requirement of at least 30 U/kg/week compared with 29% of patients who did not receive iron (P = 0.004). The average dose decrease 12 weeks after initiating iron therapy was 1763 U/week. A serum ferritin <200 ng/mL had the best positive predictive value (76%) for predicting a response to parenteral iron administration, but it still had limited clinical utility.

CONCLUSIONS

Iron deficiency commonly develops during epoetin therapy, and parenteral iron administration may result in a clinically significant reduction in epoetin dose. The use of transferrin saturation or serum ferritin as an indicator for parenteral iron administration has limited utility.

Multisite randomized controlled trials in health services research: scientific challenges and operational issues

Although well-designed randomized controlled trials (RCT) provide the strongest evidence regarding causation, only relatively recently have they been used by health services researchers to study the organization, delivery, quality, and outcomes of care. More recent yet is the extension of multisite RCTs to health services research. Such studies offer numerous methodological advantages over single-site trials: (1) enhanced external validity; (2) greater statistical power when studying conditions with a low incidence or prevalence, small event rate in the outcome (eg, mortality), and/or large variance in the outcome (eg, health care costs); and (3) rapid recruitment to provide health care organizations and policy makers with timely results. This paper begins by outlining the advantages of multisite RCTs over single-site trials. It then discusses both scientific challenges (ie, standardizing eligibility criteria, defining and standardizing the intervention, defining usual care, standardizing the data collection protocol, blinded outcome assessment, data management and analysis, measuring health care costs) and operational issues (ie, site selection, randomization procedures, patient accrual, maintaining enthusiasm, oversight) posed by multisite RCTs in health services research. Recommendations are offered to health services researchers interested in conducting such studies.

Risk-adjusted surgical outcomes

Measures of risk-adjusted outcome are particularly suited for the assessment of the quality of surgical care. The reliability of measures of quality that use surgical outcomes is enhanced by prospective data acquisition and should be adjusted for the preoperative severity of illness. Such measures should be based only on reliable and validated data, and they should apply state-of-the-art analytical methods. The risk-adjusted postoperative mortality rate is useful as a quality measure only in specialties and operations expected to have a high rate of postoperative deaths. Risk-adjusted complications are more common but are limited as a comparative measure of quality by a lack of uniform definitions and data collection mechanisms. In specialties in which the expected postoperative mortality is low, risk-adjusted functional outcomes are promising measures for the assessment of the quality of surgical care. Measures of cost and patient satisfaction should also be incorporated in systems designed to measure the quality and cost-effectiveness of surgical care.

Heritability of self-reported health

OBJECTIVE

To explore the contribution of genes and environmental factors to variation in a common measure (i.e., a five-point--excellent, very good, good, fair, and poor--Likert scale) of self-reported health.

DATA SOURCES

Data were analyzed from 4,638 male-male twin pair members of the Vietnam Era Twin (VET) Registry who responded to a 1987 health survey.

STUDY DESIGN

Varying models for the relationship between genetic and environmental influences on self-reported health were tested in an attempt to explain the relative contributions of additive genetic, shared and nonshared environmental effects, and health conditions reported since 1975 to perceived health status.

DATA COLLECTION

A mail and telephone survey of health was administered in 1987 to VET Registry twins.

PRINCIPAL FINDINGS

Variance component estimates under the best-fitting model included a 39.6 percent genetic contribution to self-reported health. In a model which included the effect of health condition, genes accounted for 32.5 percent and health condition accounted for 15.0 percent of the variance in self-reported health. The magnitude of the genetic contribution to perceived health status was not significantly different in a model with or without health condition.

CONCLUSIONS

These data suggest over one-third of the variability of self-reported health can be attributed to genes. Since perceived health status is a major predictor of morbidity, mortality, and health services utilization, future analyses should consider the role of heritable influences on traditional health services variables.

Predicting non-elective hospital readmissions: a multi-site study. Department of Veterans Affairs Cooperative Study Group on Primary Care and Readmissions

OBJECTIVE

To determine clinical and patient-centered factors predicting non-elective hospital readmissions.

DESIGN

Secondary analysis from a randomized clinical trial.

CLINICAL SETTING

Nine VA medical centers.

PARTICIPANTS

Patients discharged from the medical service with diabetes mellitus, congestive heart failure, and/or chronic obstructive pulmonary disease (COPD).

MAIN OUTCOME MEASUREMENT

Non-elective readmission within 90 days.

RESULTS

Of 1378 patients discharged, 23.3% were readmitted. After controlling for hospital and intervention status, risk of readmission was increased if the patient had more hospitalizations and emergency room visits in the prior 6 months, higher blood urea nitrogen, lower mental health function, a diagnosis of COPD, and increased satisfaction with access to emergency care assessed on the index hospitalization.

CONCLUSIONS

Both clinical and patient-centered factors identifiable at discharge are related to non-elective readmission. These factors identify high-risk patients and provide guidance for future interventions. The relationship of patient satisfaction measures to readmission deserves further study.

Efficacy of 3 commonly used hearing aid circuits: A crossover trial. NIDCD/VA Hearing Aid Clinical Trial Group

CONTEXT

Numerous studies have demonstrated that hearing aids provide significant benefit for a wide range of sensorineural hearing loss, but no carefully controlled, multicenter clinical trials comparing hearing aid efficacy have been conducted.

OBJECTIVE

To compare the benefits provided to patients with sensorineural hearing loss by 3 commonly used hearing aid circuits.

DESIGN

Double-blind, 3-period, 3-treatment crossover trial conducted from May 1996 to February 1998.

SETTING

Eight audiology laboratories at Department of Veterans Affairs medical centers across the United States.

PATIENTS

A sample of 360 patients with bilateral sensorineural hearing loss (mean age, 67.2 years; 57% male; 78.6% white).

INTERVENTION

Patients were randomly assigned to 1 of 6 sequences of linear peak clipper (PC), compression limiter (CL), and wide dynamic range compressor (WDRC) hearing aid circuits. All patients wore each of the 3 hearing aids, which were installed in identical casements, for 3 months.

MAIN OUTCOME MEASURES

Results of tests of speech recognition, sound quality, and subjective hearing aid benefit, administered at baseline and after each 3-month intervention with and without a hearing aid. At the end of the experiment, patients ranked the 3 hearing aid circuits.

RESULTS

Each circuit markedly improved speech recognition, with greater improvement observed for soft and conversationally loud speech (all 52-dB and 62-dB conditions, P</=.001). All 3 circuits significantly reduced the frequency of problems encountered in verbal communication. Some test results suggested that CL and WDRC circuits provided a significantly better listening experience than PC circuits in word recognition (P =.002), loudness (P =.003), overall liking (P =.001), aversiveness of environmental sounds (P =.02), and distortion (P =.02). In the rank-order ratings, patients preferred the CL hearing aid circuits more frequently (41.6%) than the WDRC (29.8%) and the PC (28.6%) (P =.001 for CL vs both WDRC and PC).

CONCLUSIONS

Each circuit provided significant benefit in quiet and noisy listening situations. The CL and WDRC circuits appeared to provide superior benefits compared with the PC, although the differences between them were much less than the differences between the aided vs unaided conditions. JAMA. 2000;284:1806-1813.

Effect of intensive glycemic control on microalbuminuria in type 2 diabetes. Veterans Affairs Cooperative Study on Glycemic Control and Complications in Type 2 Diabetes Feasibility Trial Investigators

OBJECTIVE

Microalbuminuria can reflect the progress of microvascular complications and may be predictive of macrovascular disease in type 2 diabetes. The effect of intensive glycemic control on microalbuminuria in patients in the U.S. who have had type 2 diabetes for several years has not previously been evaluated.

RESEARCH DESIGN AND METHODS

We randomly assigned 153 male patients to either intensive treatment (INT) (goal HbA(1c) 7.1%) or to standard treatment (ST) (goal HbA(1c) 9.1%; P = 0.001), and data were obtained during a 2-year period. Mean duration of known diabetes was 8 years, mean age of the patients was 60 years, and patients were well matched at baseline. We obtained 3-h urine samples for each patient at baseline and annually and defined microalbuminuria as an albumin:creatinine ratio of 0.03-0.30. All patients were treated with insulin and received instructions regarding diet and exercise. Hypertension and dyslipidemia were treated with similar goals in each group.

RESULTS

A total of 38% of patients had microalbuminuria at entry and were evenly assigned to both treatment groups. INT retarded the progression of microalbuminuria during the 2-year period: the changes in albumin:creatinine ratio from baseline to 2 years of INT versus ST were 0.045 vs. 0.141, respectively (P = 0.046). Retardation of progressive urinary albumin excretion was most pronounced in those patients who entered the study with microalbuminuria and were randomized to INT. Patients entering with microalbuminuria had a deterioration in creatinine clearance at 2 years regardless of the intensity of glycemic control. In the group entering without microalbuminuria, the subgroup receiving ST had a lower percentage of patients with a macrovascular event (17%) than the subgroup receiving INT (36%) (P = 0.03). Use of ACE inhibitors or calcium-channel blockers was similarly distributed among the groups.

CONCLUSIONS

Intensive glycemic control retards microalbuminuria in patients who have had type 2 diabetes for several years but may not lessen the progressive deterioration of glomerular function. Increases in macrovascular event rates in the subgroup entering without albuminuria who received INT remain unexplained but could reflect early worsening, as observed with microvascular disease in the Diabetes Control and Complications Trial.

Outcomes 15 years after valve replacement with a mechanical versus a bioprosthetic valve: final report of the Veterans Affairs randomized trial

OBJECTIVES

The goal of this study was to compare long-term survival and valve-related complications between bioprosthetic and mechanical heart valves.

BACKGROUND

Different heart valves may have different patient outcomes.

METHODS

Five hundred seventy-five patients undergoing single aortic valve replacement (AVR) or mitral valve replacement (MVR) at 13 VA medical centers were randomized to receive a bioprosthetic or mechanical valve.

RESULTS

By survival analysis at 15 years, all-cause mortality after AVR was lower with the mechanical valve versus bioprosthesis (66% vs. 79%, p = 0.02) but not after MVR. Primary valve failure occurred mainly in patients <65 years of age (bioprosthesis vs. mechanical, 26% vs. 0%, p < 0.001 for AVR and 44% vs. 4%, p = 0.0001 for MVR), and in patients > or =65 years after AVR, primary valve failure in bioprosthesis versus mechanical valve was 9 +/- 6% versus 0%, p = 0.16. Reoperation was significantly higher for bioprosthetic AVR (p = 0.004). Bleeding occurred more frequently in patients with mechanical valve. There were no statistically significant differences for other complications, including thromboembolism and all valve-related complications between the two randomized groups.

CONCLUSIONS

At 15 years, patients undergoing AVR had a better survival with a mechanical valve than with a bioprosthetic valve, largely because primary valve failure was virtually absent with mechanical valve. Primary valve failure was greater with bioprosthesis, both for AVR and MVR, and occurred at a much higher rate in those aged <65 years; in those aged > or =65 years, primary valve failure after AVR was not significantly different between bioprosthesis and mechanical valve. Reoperation was more common for AVR with bioprosthesis. Thromboembolism rates were similar in the two valve prostheses, but bleeding was more common with a mechanical valve.

Improving residents' compliance with standards of ambulatory care: results from the VA Cooperative Study on Computerized Reminders

CONTEXT

Computerized systems to remind physicians to provide appropriate care have not been widely evaluated in large numbers of patients in multiple clinical settings.

OBJECTIVE

To examine whether a computerized reminder system operating in multiple Veterans Affairs (VA) ambulatory care clinics improves resident physician compliance with standards of ambulatory care.

DESIGN, SETTING, AND PARTICIPANTS

A total of 275 resident physicians at 12 VA medical centers were randomly assigned in firms or half-day clinic blocks to either a reminder group (n = 132) or a control group (n = 143). During a 17-month study period (January 31, 1995-June 30, 1996), the residents cared for 12,989 unique patients for whom at least 1 of the studied standards of care (SOC) was applicable.

MAIN OUTCOME MEASURES

Compliance with 13 SOC, tracked using hospital databases and encounter forms completed by residents, compared between residents in the reminder group vs those in the control group.

RESULTS

Measuring compliance as the proportion of patients in compliance with all applicable SOC by their last visit during the study period, the reminder group had statistically significantly higher rates of compliance than the control group for all standards combined (58.8% vs 53.5%; odds ratio [OR], 1.24; 95% confidence interval [CI], 1.08-1.42; P =.002) and for 5 of the 13 standards examined individually. Measuring compliance as the proportion of all visits for which care was indicated in which residents provided proper care, the reminder group also had statistically significantly higher rates of compliance than the control group for all standards combined (17.9% vs 12.2%; OR, 1.57; 95% CI, 1.45-1.71; P<.001) and for 9 of the 13 standards examined individually. The benefit of reminders, however, declined throughout the course of the study, even though the reminders remained active.

CONCLUSIONS

Our data indicate that reminder systems installed at multiple sites can improve residents' compliance to multiple SOC. The benefits of such systems, however, appear to deteriorate over time. Future research needs to explore methods to better sustain the benefits of reminders. JAMA. 2000;284:1411-1416.

Two years of intensive glycemic control and left ventricular function in the Veterans Affairs Cooperative Study in Type 2 Diabetes Mellitus (VA CSDM)

OBJECTIVE

The Veterans Affairs Cooperative Study in Type 2 Diabetes Mellitus (VA CSDM) was a multicenter randomized prospective study of 153 male type 2 diabetic patients to assess the ability to sustain clinically significant glycemic separation between intensive and standard treatment arms. A trend toward an excess of combined cardiovascular events in the intensive treatment arm of this trial was reported earlier. The present analysis was done to evaluate the effect of 2 years of intensive glycemic control on the left ventricular (LV) function.

RESEARCH DESIGN AND METHODS

The patients were randomized to intensive step treatment with insulin alone or with sulfonylurea (intensive treatment arm [INT], n = 75) or to standard once-daily insulin injection (standard treatment arm [STD], n = 78) treatment. A total of 136 patients (standard treatment arm [STD], n = 70; INT, n = 66) had radionuclide ventriculography at entry and at 24 months for the assessment of LV function.

RESULTS

There was no difference in the mean LV ejection fraction (at entry: STD 57.1+/-9.51%; INT 58.1+/-8.7%; at 24 months: STD 57.3+/-10.8%, INT 59.5+/-10.7%), peak filling rate (at entry: STD 2.6+/-0.7 end diastolic volume per second, INT 2.4+/-0.8 end diastolic volume per second; at 24 months: STD 2.7+/-1.0 end diastolic volume per second, INT 2.5+/-0.7 end diastolic volume per second), or time to peak filling rate (at entry: STD 195.3+/-69.5 ms, INT 185.6 +/-62.4 ms; at 24 months: STD 182.6+/-64.8 ms, INT 179.2+/-61.2 ms) between the 2 treatment arms. A subgroup analysis of 104 patients (STD, n = 53; INT, n = 51) that omitted individuals with intervening cardiac events/revascularization or a change in cardioactive medications also showed no difference in the LV function at entry and at 24 months between the 2 groups. Abnormal LV ejection fraction at baseline predicted cardiac events (interval between cardiac beats [RR] = 2.5).

CONCLUSIONS

Two years of intensive glycemic control does not affect the LV systolic or diastolic function in patients with type 2 diabetes.

Multifactorial risk index for predicting postoperative respiratory failure in men after major noncardiac surgery. The National Veterans Administration Surgical Quality Improvement Program

OBJECTIVE

To develop and validate a preoperative risk index for predicting postoperative respiratory failure (PRF).

SUMMARY BACKGROUND DATA

Respiratory failure is an important postoperative complication.

METHOD

Based on a prospective cohort study, cases from 44 Veterans Affairs Medical Centers (n = 81,719) were used to develop the models. Cases from 132 Veterans Affairs Medical Centers (n = 99,390) were used as a validation sample. PRF was defined as mechanical ventilation for more than 48 hours after surgery or reintubation and mechanical ventilation after postoperative extubation. Ventilator-dependent, comatose, do not resuscitate, and female patients were excluded.

RESULTS

PRF developed in 2,746 patients (3.4%). The respiratory failure risk index was developed from a simplified logistic regression model and included abdominal aortic aneurysm repair, thoracic surgery, neurosurgery, upper abdominal surgery, peripheral vascular surgery, neck surgery, emergency surgery, albumin level less than 30 g/L, blood urea nitrogen level more than 30 mg/dL, dependent functional status, chronic obstructive pulmonary disease, and age.

CONCLUSIONS

The respiratory failure risk index is a validated model for identifying patients at risk for developing PRF and may be useful for guiding perioperative respiratory care.

Mild renal failure is associated with adverse outcome after cardiac valve surgery

The present study was performed to ascertain whether the presence of mild renal failure (defined as a serum creatinine concentration of 1. 5 to 3.0 mg/dL) is an independent risk factor for adverse outcome after cardiac valve surgery. An extensive set of preoperative and postoperative data was collected in 834 prospectively evaluated patients undergoing cardiac valve surgery at 14 Veterans Affairs Medical Centers. Univariate and multivariable analyses were performed to determine whether an independent association of mild renal dysfunction with adverse outcomes was present. Patients with mild renal failure had significantly greater 30-day mortality rates (P = 0.001; 16% versus 6%) and frequency of postoperative bleeding (P = 0.023; 16% versus 8%), respiratory complications (P = 0.02, 29% versus 16%), and cardiac complications (P = 0.002; 18% versus 7%) than patients with normal renal function (serum creatinine <1.5 mg/dL) when controlling for multiple other variables. The presence of a serum creatinine concentration of 1.5 to 3.0 mg/dL is significantly and independently associated with adverse outcomes after cardiac valve surgery.

Risk factors for morbidity and mortality after colectomy for colon cancer

PURPOSE

Comorbid conditions affect the risk of adverse outcomes after surgery, but the magnitude of risk has not previously been quantified using multivariate statistical methods and prospectively collected data. Identifying factors that predict results of surgical procedures would be valuable in assessing the quality of surgical care. This study was performed to define risk factors that predict adverse events after colectomy for cancer in Department of Veterans Affairs Medical Centers.

METHODS

The National Veterans Affairs Surgical Quality Improvement Program contains prospectively collected and extensively validated data on more than 415,000 surgical operations. All patients undergoing colectomy for colon cancer from 1991 to 1995 who were registered in the National Veterans Affairs Surgical Quality Improvement Program database were selected for study. Independent variables examined included 68 preoperative and 12 intraoperative clinical risk factors; dependent variables were 21 specific adverse outcomes. Stepwise logistic regression analysis was used to construct models predicting the 30-day mortality rate and 30-day morbidity rates for each of the ten most frequent complications.

RESULTS

A total of 5,853 patients were identified; 4,711 (80 percent) underwent resection and primary anastomosis. One or more complications were observed in 1,639 of 5,853 (28 percent) patients. Prolonged ileus (439/5,853; 7.5 percent), pneumonia (364/5,853; 6.2 percent), failure to wean from the ventilator (334/5,853; 5.7 percent), and urinary tract infection (292/5,853; 5 percent) were the most frequent complications. The 30-day mortality rate was 5.7 percent (335/5,853). For most complications, 30-day in-hospital mortality rates were significantly higher for patients with a complication than for those without. Thirty-day mortality rates exceeded 50 percent if postoperative coma, cardiac arrest, a pre-existing vascular graft prosthesis that failed after colectomy, renal failure, pulmonary embolism, or progressive renal insufficiency occurred. Preoperative factors that predicted a high risk of 30-day mortality included ascites, serum sodium >145 mg/dl, "do not resuscitate" status before surgery, American Society of Anesthesiologists classes III and IV OR V, and low serum albumin.

CONCLUSIONS

Mortality rates after colectomy in Veterans Affairs hospitals are comparable with those reported in other large studies. Ascites, hypernatremia, do not resuscitate status before surgery, and American Society of Anesthesiologists classes III and IV OR V were strongly predictive of perioperative death. Clinical trials to decrease the complication rate after colectomy for colon cancer should focus on these risk factors.

The effects of intensive glycemic control on neuropathy in the VA cooperative study on type II diabetes mellitus (VA CSDM)

To determine whether a difference in HbA(1c) could be safely sustained between a standard therapy (STD) arm and an intensive therapy (INT) arm, while maintaining HbA(1c) levels in both arms within a range acceptable in community practice. The effects of intensive treatment on various parameters were studied in this feasibility trial. We report here the results of 24 months of INT on peripheral and autonomic neuropathy.A prospective trial was conducted in five medical centers in 153 men of 60 +/- 6 years of age who had a known diagnosis of diabetes for 7.8 +/- 4 years. They were randomly assigned to a standard insulin treatment group (one morning injection per day) or to an intensive therapy group designed to attain near-normal glycemia and a clinically significant separation of glycohemoglobin from the standard arm. A four-step plan was used in the intensive therapy group along with daily self-monitoring of glucose: (1) an evening insulin injection, (2) the same injection adding daytime glipizide, (3) two injections of insulin alone, and (4) multiple daily injections. Peripheral neuropathy was diagnosed clinically by a history and physical examination, and by abnormal autonomic neuropathy Valsalva ratio (VR < 1.2) and RR variation (RRV < 10). An average HbA(1c) separation of 2.07% was achieved with INT, having HbA(1c) at or below 7.3% (p = 0. 001 versus STD). Baseline prevalence of peripheral neuropathy was 53% in STD, and 48% in INT. By 24 months, the prevalence increased to 69% in STD (p = 0.005 versus baseline), and to 64% in INT (p = 0. 008 versus baseline, but no different than STD). Though INT did not reverse all elements of peripheral neuropathy, there was a decreased prevalence of cranial neuropathy (p = 0.053 versus STD) and more frequent preservation of touch sensation in the upper extremities (p = 0.03 versus STD) in INT. At baseline, an abnormal Valsalva ratio and/or RR variation was seen in 38% of STD and 31% of INT. By 24 months in STD, the prevalence rose to 55% (p = 0.0067 versus baseline), and in INT, to 48% (p = 0.012 versus baseline and no different from STD). The prevalence of erectile dysfunction increased from 53% at baseline to 73% at 2 years, p = 0.002 in STD, and from 51% to 73% at 2 years (p = 0.003 versus baseline) and no different from STD. There was no change in the frequency of abnormal gastrointestinal or sweating symptoms. Our conclusion was that 2 years of meticulous glycemic control did not decrease overall prevalence of peripheral or autonomic neuropathy. In fact, the prevalence rose equivalently and significantly in both treatment arms. There was some benefit, however, in decreased frequency of cranial neuropathy and better preservation of touch sensation in INT.

Stroke in cardiac surgical patients: determinants and outcome

BACKGROUND

Despite improving outcomes in cardiac surgical patients, stroke continues to remain a major complication. Few prospective studies are available on postoperative stroke. The present study was conducted to elucidate the incidence and predictors of stroke in a large group of cardiac surgical patients.

METHODS AND RESULTS

Prospective data collected on 4,941 patients undergoing cardiac surgery were subjected to univariate and logistic regression analyses (98.4% men; 72% older than 60 years; 9.1% with history of prior stroke; 80.4% underwent isolated coronary artery bypass grafting). Stroke predictors include history of stroke and hypertension, older age, systolic hypertension, bronchodilator and diuretic use, high serum creatinine, surgical priority, great vessel repair, use of inotropic agents after cardiopulmonary bypass, and total cardiopulmonary bypass time (p < 0.05 for all comparisons). Median intensive care unit and hospital stays were longer, and hospital mortality and 6-month mortality were higher for patients with stroke (p < 0.001).

CONCLUSIONS

Stroke after cardiac surgical procedures is a morbid event. Identification of predictors and development of strategies to modify these factors should lead to a lower incidence of stroke.

Racial variation in the use of laparoscopic cholecystectomy in the Department of Veterans Affairs medical system

BACKGROUND

While studies have found racial differences in the rates of use of established invasive cardiac and cerebrovascular procedures, no study has evaluated racial variation in the rates of adoption of new surgical procedures. For patients undergoing laparoscopic cholecystectomy, the procedure represents a new and safe option that shortens the duration of postoperative hospitalization by almost one week. In this study, we evaluated whether, in the equal access Veterans Affairs (VA) medical system, the rate of adoption of this procedure and improvements in the duration of postoperative hospitalization differed between African-American and Caucasian patients.

STUDY DESIGN

Data were obtained from two sources-administrative claims files and prospectively compiled dinical data from medical records and patient interviews. In both data sets, frequency of use, length of stay, and outcomes for African-American and Caucasian patients undergoing minimally invasive and open gallbladder surgery were analyzed for the first four years of use of the procedure in the VA system (1992 to 1995).

RESULTS

Analyses based on claims files indicated that, after adjustment for potentially confounding variables, African-American patients who underwent cholecystectomy in VA medical centers were 25% less likely to undergo a minimally invasive cholecystectomy during the first 4 years of use of the new procedure (adjusted odds ratio, 0.74; 95% confidence interval, 0.66-0.83). Shortening of the average postoperative length of stay from 9 days or more in the prelaparoscopic era to less than 4.5 days for patients undergoing the laparoscopic procedure occurred in the first year for Caucasian patients, but did not occur until the fourth year for African-American patients (p<0.001). The overall difference in postoperative length of stay between African-American and Caucasian patients more than doubled from 1.7 days before introduction of laparoscopic cholecystectomy to 3.8 days in the fourth year. In comparison, analyses based on nurse-compiled clinical data indicated that, after adjustment for relevant clinical factors, racial variations in the rate of laparoscopic surgery were even larger (adjusted odds ratio for laparoscopic versus open cholecystectomy for African-American versus Caucasian veterans, 0.68; 95% confidence interval, 0.55-0.84).

CONCLUSIONS

Compared to Caucasian patients, African-American patients who underwent cholecystectomy in VA medical centers had an approximately 25% to 32% lower likelihood of undergoing minimally invasive cholecystectomy procedures. The differences in rates of adoption of laparoscopic surgery did not appear to be from more comorbid illnesses among African-American patients. African-American and Caucasian veterans may differ in their preference for new surgical procedures like laparoscopic cholecystectomy. Conversely, VA physicians may have been less likely to recommend laparoscopic cholecystectomies to African-American patients.

Prophylactic coronary artery revascularization for elective vascular surgery: study design. Veterans Affairs Cooperative Study Group on Coronary Artery Revascularization Prophylaxis for Elective Vascular Surgery

This article describes the design of an ongoing randomized trial intended to test whether patients who require elective vascular surgery would benefit from preoperative coronary artery revascularization prior to the vascular procedure. The primary objective is to determine whether coronary artery revascularization reduces long-term mortality (mean 3.5 years) in patients undergoing vascular surgery. The study design calls for 620 patients to be randomized and followed for a mean of 3.5 years following vascular surgery. Secondary endpoints include measures of quality of life and cost-effectiveness. Patients with coronary artery disease in need of an elective vascular operation are considered candidates for the study. Anatomic exclusion criteria include ejection fraction <20%, severe aortic stenosis (valve area <1.0 cm2), left main stenosis > or =50%, nonobstructive coronary artery disease (stenosis <70%), and coronary arteries that are not amenable to revascularization. Prior to the vascular surgery, the trial randomizes eligible patients to coronary artery revascularization (either bypass surgery or angioplasty) versus medical therapy. The trial stratifies the randomization by hospital and type of vascular surgery (intraabdominal versus infrainguinal) because of differences in long-term prognosis in those patients. A 1-year feasibility trial involving five Veterans Affairs (VA) medical centers of variable vascular surgical loads has been completed. The results showed that over 90% of expected patients could be randomized. As a result, a larger VA Cooperative Study involving 18 centers will begin recruitment of patients. The findings should help determine the best strategy for managing patients with coronary artery disease in need of elective vascular surgery.

Prognostic models of thirty-day mortality and morbidity after major pulmonary resection

BACKGROUND

A part of the prospective, multi-institutional National Veterans Affairs Surgical Quality Improvement Program was developed to predict 30-day mortality and morbidity for patients undergoing a major pulmonary resection.

METHODS

Perioperative data were acquired from 194,319 noncardiac surgical operations at 123 Veterans Affairs Medical Centers between October 1, 1991, and August 31, 1995. Current Procedural Terminology code-based analysis was undertaken for major pulmonary resections (lobectomy and pneumonectomy). Preoperative, intraoperative, and outcome variables were collected. The 30-day mortality and morbidity models were developed by means of multivariable stepwise logistic regression with the preoperative and intraoperative variables used as independent predictors of outcome.

RESULTS

A total of 3516 patients (mean age 64 9 years) underwent either lobectomy (n = 2949) or pneumonectomy (n = 567). Thirty-day mortality was 4.0% for lobectomy (119/2949) and 11.5% for pneumonectomy (65/567). The preoperative predictors of 30-day mortality were albumin, do not resuscitate status, transfusion of more than 4 units, age, disseminated cancer, impaired sensorium, prothrombin time more than 12 seconds, type of operation, and dyspnea. When the intraoperative variables were considered, intraoperative blood loss was added to the preoperative model. In the presence of these intraoperative variables in the model, do not resuscitate status and prothrombin time more than 12 seconds were only marginally significant. Thirty-day morbidity, defined as the presence of 1 or more of the 21 predefined complications, was 23.8% for lobectomy (703/2949) and 25.7% for pneumonectomy (146/567). In multivariable models, independent preoperative predictors (P <.05) of 30-day morbidity were age, weight loss greater than 10% in the 6 months before surgery, history of chronic obstructive pulmonary disease, transfusion of more than 4 units, albumin, hemiplegia, smoking, and dyspnea. When intraoperative variables were added to the preoperative model, the duration of operation time and intraoperative transfusions were included in the model and albumin became marginally significant.

CONCLUSIONS

This analysis identifies independent patient risk factors that are associated with 30-day mortality and morbidity for patients undergoing a major pulmonary resection. This series provides an initial risk-adjustment model for major pulmonary resections. Future refinements will allow comparative assessment of surgical outcomes and quality of care at many institutions.

Health-related quality of life as a predictor of mortality following coronary artery bypass graft surgery. Participants of the Department of Veterans Affairs Cooperative Study Group on Processes, Structures, and Outcomes of Care in Cardiac Surgery

CONTEXT

Health-related quality of life has not been evaluated as a predictor of mortality following coronary artery bypass graft (CABG) surgery. Evaluation of health status as a mortality predictor may be useful for preoperative risk stratification.

OBJECTIVE

To determine whether the Physical and Mental Component Summary scores from the preoperative Short-Form 36 (SF-36) health status survey predict mortality following CABG surgery after adjustment for known clinical risk variables.

DESIGN

Prospective cohort study conducted between September 1992 and December 1996.

SETTING

Fourteen Veterans Affairs hospitals.

PATIENTS

Of the 3956 patients undergoing CABG surgery only and who were enrolled in the Processes, Structures, and Outcomes of Care in Cardiac Surgery study, the 2480 who completed a preoperative SF-36.

MAIN OUTCOME MEASURE

All-cause mortality within 180 days after surgery.

RESULTS

A total of 117 deaths (4.7%) occurred within 180 days of CABG surgery. The Physical Component Summary of the preoperative SF-36 was a statistically significant risk factor for 6-month mortality after adjustment for known clinical risk factors for mortality following CABG surgery. In multivariate analysis, a 10-point lower SF-36 Physical Component Summary score had an odds ratio (OR) of 1.39 (95% confidence interval [CI], 1.11-1.77; P=.006) for predicting mortality. The SF-36 Mental Component Summary score was not associated with 6-month mortality in multivariate analyses (OR, 1.09; 95% CI, 0.92-1.29; P=.31).

CONCLUSIONS

The Physical Component Summary score from the preoperative SF-36 is an independent risk factor for mortality following CABG surgery. The baseline Mental Component Summary score does not appear to be predictive of mortality. Preoperative patient self-report of the physical component of health status may be helpful for risk stratification and clinical decision making for patients undergoing CABG surgery.

Renal failure predisposes patients to adverse outcome after coronary artery bypass surgery. VA Cooperative Study #5

BACKGROUND

More than 600,000 coronary artery bypass graft (CABG) procedures are done annually in the United States. Some data indicate that 10 to 20% of patients who are undergoing a CABG procedure have a serum creatinine of more than 1.5 mg/dl. There are few data on the impact of a mild increase in serum creatinine concentration on CABG outcome.

METHODS

We analyzed a Veterans Affairs database obtained prospectively from 1992 through 1996 at 14 of 43 centers performing heart surgery. We compared the outcome after CABG in patients with a baseline serum creatinine of less than 1.5 mg/dl (median 1.1 mg/dl, N = 3271) to patients with a baseline serum creatinine of 1.5 to 3.0 mg/dl (median 1.7, N = 631).

RESULTS

Univariate analysis revealed that patients with a serum creatinine of 1.5 to 3.0 mg/dl had a higher 30-day mortality (7% vs. 3%, P < 0.001) requirement for prolonged mechanical ventilation (15% vs. 8%, P = 0.001), stroke (7% vs. 2%, P < 0.001), renal failure requiring dialysis at discharge (3% vs. 1%, P < 0.001), and bleeding complications (8% vs. 3%, P < 0.001) than patients with a baseline serum creatinine of less than 1.5 mg/dl. Multiple logistic regression analyses found that patients with a baseline serum creatinine of less than 1.5 mg/dl had significantly lower (P < 0.02) 30-day mortality and postoperative bleeding and ventilatory complications than patients with a serum creatinine of 1.5 to 3.0 mg/dl when controlling for all other variables.

CONCLUSION

These results demonstrate that mild renal failure is an independent risk factor for adverse outcome after CABG.

The Veterans Affairs Implantable Insulin Pump Study: effect on cardiovascular risk factors

OBJECTIVE

To determine whether implantable insulin pump (IIP) and multiple-dose insulin (MDI) therapy have different effects on cardiovascular risk factors in insulin-requiring patients with type 2 diabetes.

RESEARCH DESIGN AND METHODS

A randomized clinical trial was conducted at seven Veterans Affairs medical centers in 121 male patients with type 2 diabetes between the ages of 40 and 69 years receiving at least one injection of insulin per day and with HbA1c, levels of > or =8% at baseline. Weights, blood pressures, insulin use, and glucose monitoring data were obtained at each visit. Lipid levels were obtained at 0, 4, 8, and 12 months, and free and total insulin levels were obtained at 0, 6, and 12 months. All medications being taken were recorded at each visit.

RESULTS

No difference in absolute blood pressure, neither systolic nor diastolic, was seen between patients receiving MDI or IIP therapy, but significantly more MDI patients required anti-hypertensive medications. When blood pressure was modeled against weight and time, IIP therapy was significantly better than MDI therapy for systolic blood pressure in patients with BMI <33 and for diastolic blood pressure in patients with BMI >34 kg/m2. Total cholesterol levels decreased in the overall sample, but IIP patients exhibited significantly higher levels than MDI patients. Triglyceride levels increased over time for both groups, with IIP patients having significantly higher levels than patients in the MDI group. BMI was a significant predictor of, and inversely proportional to, HDL cholesterol level. No difference in lipid-lowering drug therapy was seen between the two groups. Free insulin and insulin antibodies tended to decrease in the IIP group as compared with the MDI group. C-peptide levels decreased in both groups.

CONCLUSIONS

IIP therapy in insulin-requiring patients with type 2 diabetes has advantages over MDI therapy in decreasing the requirement for antihypertensive therapy and for decreasing total and free insulin and insulin antibodies. Both therapies reduce total cholesterol and C-peptide levels.

Subcutaneous compared with intravenous epoetin in patients receiving hemodialysis. Department of Veterans Affairs Cooperative Study Group on Erythropoietin in Hemodialysis Patients

BACKGROUND

Several studies have suggested that if recombinant human erythropoietin (epoetin) is administered subcutaneously rather than intravenously, a lower dose may be sufficient to maintain the hematocrit at a given level.

METHODS

In a randomized, unblinded trial conducted at 24 hemodialysis units at Veterans Affairs medical centers, we assigned 208 patients who were receiving long-term hemodialysis and epoetin therapy to treatment with either subcutaneous or intravenous epoetin. The dose was initially reduced until the hematocrit was below 30 percent and then was gradually increased to a level that would maintain the hematocrit in the range of 30 to 33 percent for 26 weeks. We compared the average doses in the 26-week maintenance phase and the discomfort associated with the two routes of administration.

RESULTS

For the 107 patients treated by the subcutaneous route, the average weekly dose of epoetin during the maintenance phase was 32 percent less than that for the 101 patients treated by the intravenous route (mean [+/-SD], 95.1+/-75.0 vs. 140.3+/-88.5 U per kilogram of body weight per week; P<0.001). Only one patient in the subcutaneous-therapy group withdrew from the study because of pain at the injection site, and 86 percent rated the pain associated with subcutaneous administration as ranging from absent to mild.

CONCLUSIONS

In patients receiving hemodialysis, subcutaneous administration of epoetin can maintain the hematocrit in a desired target range, with an average weekly dose of epoetin that is lower than with intravenous administration.

Outcome after proctectomy for rectal cancer in Department of Veterans Affairs Hospitals: a report from the National Surgical Quality Improvement Program

OBJECTIVE

To define risk factors that predict adverse outcomes after proctectomy for cancer in Department of Veterans Affairs Medical Centers.

SUMMARY BACKGROUND DATA

Accurate presurgical assessment of the risk of perioperative complications and death is important in planning surgical therapy.

METHODS

The National VA Surgical Quality Improvement Program contains prospectively collected and extensively validated data on >287,000 patients. All patients undergoing proctectomy for rectal cancer from 1991 to 1995 who were registered in this data base were selected for study. Independent variables examined included 68 presurgical and 12 intraoperative clinical risk factors; dependent variables were 21 specific adverse outcomes. Stepwise logistic regression analysis was used to construct models predicting 30-day morbidity rates for each of the 10 most common complications and the 30-day mortality rate.

RESULTS

Five hundred ninety-one patients were identified; 467 (79%) underwent abdominoperineal resection and 124 (21%) were treated with sphincter-saving procedures. Thirty percent of patients had one or more complications after proctectomy. Prolonged ileus, urinary tract infection, pneumonia, and deep wound infection were the most frequently reported complications. The 30-day mortality rate was 3.2% (19 deaths). For most complications, 30-day mortality rates were significantly higher for patients with complications than for those without. Thirty-day mortality rates for several complications exceeded 50%: cardiac arrest requiring cardiopulmonary resuscitation, deep venous thrombosis or thrombophlebitis, coma lasting >24 hours, acute renal failure, cerebrovascular accident, and pulmonary embolism. Four presurgical factors predicted a high risk of 30-day mortality in the logistic regression analysis: elevated blood urea nitrogen level, impaired sensorium, low serum albumin concentration, and partial thromboplastin time < or =25 seconds.

CONCLUSIONS

Mortality rates after proctectomy in VA hospitals are comparable to those reported in other large series. Most postsurgical complications are associated with an increased 30-day mortality rate. Elevated presurgical blood urea nitrogen level, impaired sensorium, low serum albumin concentration, and partial thromboplastin time < or =25 seconds predict a high risk of 30-day mortality.

Modelling ordinal responses from co-twin control studies

The co-twin control design has been widely used in studying the effects of environmental factors on the development of diseases. For binary outcomes that arise from co-twin control studies, the conditional likelihood method is commonly used. This approach, however, does not readily extend to ordinal response data because the standard conditional likelihood does not exist for cumulative logit or proportional odds models. In this paper, we investigate the applicability of the random-effects and GEE approaches in analysing ordinal response data from co-twin control studies. Using both approaches, we re-analyse data from a co-twin control study of the impact of military services during the Vietnam era on post-traumatic stress disorders (PTSD). The ordinal models have considerably increased power in detecting the effects of exposure when compared to the analyses using a dichotomized response. We discuss the interpretation of the estimates from GEE and random-effect models in the context of the twin data.

Long-term reliability and validity of alcoholism diagnoses and symptoms in a large national telephone interview survey

The long-term reliability and validity of telephone lay interview assessments of alcoholism were examined in the context of a large national community-based survey of over 8,000 male Vietnam era veterans. A subsample of 146 men was interviewed twice by telephone using the same structured interview an average of 15 months apart to evaluate the long-term reliability of alcoholism symptoms and diagnoses. In addition, a search of Department of Veterans Affairs patient treatment files of inpatient hospitalizations between 1970 and 1993 yielded a subsample of 89 interviewed men with a past discharge diagnosis of alcohol dependence. The test-retest reliability of alcohol abuse and alcohol dependence diagnoses was good, with kappa coefficients of 0.74 and 0.61, respectively. The reliability of individual alcoholism symptoms was fair to good, with kappas of 0.46 to 0.67. Ninety-six percent of individuals identified by Department of Veterans Affairs patient treatment files as having an alcohol dependence diagnosis were correctly diagnosed by the telephone interview. The results of the present study provide additional evidence for the long-term reliability and validity of lifetime alcoholism diagnoses, and suggest that the reliability and validity of telephone interview assessments of alcoholism are as good as that of an in-person interview. Telephone administration of structured psychiatric interviews appears to be an attractive alternative to in-person interviewing for gathering information about alcoholism and alcohol-related problems.

Ethnic differences in the glycemic response to exogenous insulin treatment in the Veterans Affairs Cooperative Study in Type 2 Diabetes Mellitus (VA CSDM)

OBJECTIVE

The Veterans Affairs Cooperative Study in Type 2 Diabetes Mellitus was conducted in NIDDM patients to determine if a significant difference in HbA1c could be achieved between groups receiving standard and intensive treatment. We observed differences in the response to exogenous insulin between African-Americans and other intensively treated patients. Therefore, we assessed the variations of response and correlated factors that might explain such differences.

RESEARCH DESIGN AND METHODS

One hundred fifty-three men aged 40-69 years with NIDDM for < or = 15 years were randomized to either the standard therapy (n = 78) or the intensive therapy (n = 75) arm. Of the 75 patients in the intensive therapy group, 57 completed the study on insulin therapy alone. Of these, 18 were African-Americans and 39 were non-African-Americans. We conducted an analysis of the data collected to determine differences in baseline characteristics, glycemic response, insulin requirement, body weight, exercise, and basal C-peptide level, factors that may explain a difference in response to insulin therapy.

RESULTS

Glycemic control improved in all patients with intensive insulin therapy. African-Americans achieved a greater improvement in HbA1c compared with non-African-Americans with a similar increment in insulin. This difference could not be explained by differences in body weight, activity, concomitant use of other medicines, or insulin-secretory capacity of the pancreas.

CONCLUSIONS

We conclude that ethnic differences may exist in the response to insulin therapy. A knowledge of such differences may aid in achieving good glycemic control, especially since minorities have a greater prevalence of and burden from the microvascular complications of diabetes.

Cooperative studies in health services research in the Department of Veterans Affairs

The Department of Veterans Affairs, through its Cooperative Studies Program, has a long history of conducting large-scale, multihospital biomedical clinical trials. The agency's Health Services Research and Development Service, although newer, has a distinguished record of mainly single-site research into the organization, delivery, and financing of health services. In 1990, a joint program was initiated to conduct multicenter studies in health services research. This article describes the studies developed in the new program and the research design issues encountered in planning them. Identification of the patient population, specification and measurement of the intervention, and description of the control group, as well as attention to the unit of randomization and analysis, outcome variables and choice of effect size, data quality, and ethical considerations are among the important issues related to the design of these studies and future studies in health services.

Response to intensive therapy steps and to glipizide dose in combination with insulin in type 2 diabetes. VA feasibility study on glycemic control and complications (VA CSDM)

OBJECTIVE

The feasibility study for the VA Cooperative Study on Glycemic Control and Complications in Type 2 Diabetes (VA CSDM) prospectively studied 153 insulin-requiring type 2 diabetes patients, randomized between an intensively treated arm and a standard treatment arm during a mean follow-up of 27 months. The glycemic response to each of the progressive, sequential phases of insulin treatment was assessed, along with the incidence of hypoglycemic reactions and the relative efficacy of different doses of glipizide in combination with fixed doses of insulin.

RESEARCH DESIGN AND METHODS

Five medical centers participated; half of the patients were assigned to the intensive treatment arm aiming for normal HbA1c levels. Age of patients was 60 +/- 6 years, duration of diabetes 8 +/- 3 years, and BMI 30.7 +/- 4 kg/m2. A four-step management technique was used, with patients moving to the next step if the operational goals were not met: Phase I, evening intermediate or long-acting insulin; phase II, added day-time glipizide; phase III, two injections of insulin alone; and phase IV, multiple daily insulin injections. Home glucose monitoring measurements were done twice daily and at 3:00 A.M. once a week. Hypoglycemic reactions and home glucose monitoring results were recorded and counted in each of the treatment phases.

RESULTS

Baseline HbA1c was 9.3 +/- 1.8%, and fasting plus serum glucose was 11.4 +/- 3.3 mmol/1. Fasting serum glucose fell to near normal in phase I, and remained so in the other treatment phases. An HbA1c separation of 2.1% between the arms was maintained during the course of the study, while the intensive arm kept HbA1c levels below 7.3% (P = 0.001). Most of the decrease in HbA1c occurred with one injection of insulin alone (phase I, -1.4%) or adding day-time glipizide (phase II, -1.9% compared with baseline). HbA1c did not decrease further after substituting two injections of insulin alone, with twice the insulin dose. Multiple daily injections resulted in an additional HbA1c fall (-2.4% compared with baseline). However, two-thirds of the patients were still on one or two injections a day at the end of the study. Changes in home glucose monitoring levels paralleled those of the HbA1c, as did the increments in number of reported hypoglycemic reactions, virtually all either "mild" or "moderate" in character. For the combination of glipizide and insulin (phase II), the only significant effect was obtained with daily doses up to 10 mg a day; there were no significant additional benefits with up to fourfold higher daily doses, and HbA1c levels had an upward trend with doses > 20 mg/day.

CONCLUSIONS

A simple regime of a single injection of insulin, alone or with glipizide, seemed sufficient to obtain clinically acceptable levels of HbA1c for most obese, insulin-requiring type 2 diabetes patients. Further decrease of HbA1c demanded multiple daily injections at the expense of doubling the insulin dose and the rate of hypoglycemic events. In combination therapy, doses of glipizide > 20 mg/day offered no additional benefit.

Metachronous bilateral Wilms' tumor: the importance of time interval to the development of a second tumor

BACKGROUND

Information regarding survival and prognostic factors in children with metachronous bilateral Wilms' tumor is limited.

METHODS

A literature search of English language articles from 1950-1996 was performed. A total of 108 of 123 children with metachronous bilateral Wilms' tumor from 30 studies were evaluable for analysis. The children were classified according to the time interval to the development of a contralateral Wilms' tumor (< 18 months vs. > or = 18 months).

RESULTS

Kaplan-Meier analysis rates of overall survival for metachronous bilateral Wilms' tumor were 49.1% and 47.2% at 5 and 10 years, respectively. The median time interval to the development of a second tumor was 23.1 months. Of 106 children, 102 (96.2%) had a metachronous presentation of Wilms' tumor by 5 years. In children ages < 2 years and children ages 2-5 years, 95.2% and 93.9%, respectively, of contralateral tumors appeared within 60 months. For children ages > or = 5 years, all contralateral kidney tumors appeared by 54 months. Analysis of overall survival of patients with a time interval of < 18 months and > or = 18 months showed a 10-year survival of 39.6% and 55.2%, respectively (P = 0.024, log rank test).

CONCLUSIONS

Children with a metachronous bilateral Wilms' tumor who developed a contralateral tumor > or = 18 months from the initial diagnosis of Wilms' tumor had a better overall survival than children with a time interval of < 18 months. Children ages < 2 years at the initial diagnosis of Wilms' tumor did not have a longer period of risk for developing a contralateral kidney tumor than those ages > or = 2 years. Screening by abdominal ultrasound of the contralateral kidney for > 5 years after initial diagnosis of Wilms' tumor may not be necessary because > 95% of children had a time interval to the development of a second tumor of < or = 60 months.

Growing use of laparoscopic cholecystectomy in the national Veterans Affairs Surgical Risk Study: effects on volume, patient selection, and selected outcomes

OBJECTIVE

To study the introduction of laparoscopic cholecystectomy to the 43 tertiary-care university-affiliated Veterans Administration medical centers (VAMCs) participating in the National Veterans Affairs Surgical Risk Study from October 1991 through December 1993.

SUMMARY BACKGROUND DATA

Previous studies in the private sector have documented growth in the number of cholecystectomies and falling clinical thresholds for cholecystectomy with the introduction of laparoscopic cholecystectomy.

METHODS

The following were analyzed for changes over time: measures of patient preoperative risk, complexity of surgery, severity of biliary disease, numbers of procedures, postoperative length of stay, and 30-day postoperative mortality and general complication rates.

RESULTS

The number of cholecystectomies performed laparoscopically increased, but the total number of cholecystectomies performed remained stable over time. The proportion of patients with acute cholecystitis, emergent cholecystectomies, and technically complex cholecystectomies did not change or increased slightly over time. Adjusted odds for postoperative general complications were lower for laparoscopic than for open cholecystectomy, but 30-day postoperative mortality and general complication rates for all cholecystectomies remained constant over time. Postoperative length of stay for all cholecystectomies fell significantly. Implementation rates of laparoscopic cholecystectomy varied widely between hospitals. Laparoscopic cholecystectomy was adopted more slowly and used in a lower percentage of cholecystectomies than in non-VA settings.

CONCLUSIONS

In contrast to non-VA studies showing increases in overall cholecystectomy volume since the introduction of laparoscopic cholecystectomy, these VAMCs implemented laparoscopic cholecystectomy without growth in cholecystectomies or a change in the clinical threshold for cholecystectomy. Laparoscopic cholecystectomy was associated with better outcomes, but its introduction in the setting of stable cholecystectomy volume and biliary disease case mix did not change postoperative mortality and complication rates. The stable cholecystectomy volume and biliary disease case mix, slower adoption, and lower use of laparoscopic cholecystectomy contrast with previous reports and may result from differences in patients and organization and financing of VA versus non-VA settings.

Atrial fibrillation after cardiac surgery: a major morbid event?

OBJECTIVE

The purpose of the study was to investigate the incidence, predictors, morbidity, and mortality associated with postoperative atrial fibrillation (AF) and its impact on intensive care unit (ICU) and postoperative hospital stay in patients undergoing cardiac surgery in the Department of Veterans Affairs (VA).

SUMMARY BACKGROUND DATA

Postoperative AF after open cardiac surgery is rather common. The etiology of this arrhythmia and factors responsible for its genesis are unclear, and its impact on postoperative surgical outcomes remains controversial. The purpose of this special substudy was to elucidate the incidence of postoperative AF and the factors associated with its development, as well as the impact of AF on surgical outcome.

METHODS

The study population consisted of 3855 patients who underwent open cardiac surgery between September 1993 and December 1996 at 14 VA Medical Centers. Three hundred twenty-nine additional patients were excluded because of lack of complete data or presence of AF before surgery, and 3794 (98.4%) were male with a mean age of 63.7+/-9.6 years. Operations included coronary artery bypass grafting (CABG) (3126, 81%), CABG + AVR (aortic valve replacement) (228, 5.9%), CABG + MVR (mitral valve replacement) (35, 0.9%), AVR (231, 6%), MVR (41, 1.06%), CABG + others (95, 2.46%), and others (99, 2.5%). The incidence of postoperative AF was 29.6%. Multivariate logistic regression analysis of factors found significant on univariate analysis showed the following predictors of postoperative AF: preoperative patient risk predictors: advancing age (odds ratio [OR] 1.61, 95% confidence interval [CI] 1.48-1.75, p < 0.001), chronic obstructive pulmonary disease (OR 1.37, 95% CI 1.12-1.66, p < 0.001), use of digoxin within 2 weeks before surgery (OR 1.37, 95% CI 1.10-1.70, p < 0.003), low resting pulse rate <80 (OR 1.26, 95% CI 1.06-1.51, p < 0.009), high resting systolic blood pressure >120 (OR 1.19, 95% CI 1.02-1.40, p < 0.026), intraoperative process of care predictors: cardiac venting via right superior pulmonary vein (OR 1.42, 95% CI 1.21-1.67, p < 0.0001), mitral valve repair (OR 2.86, 95% CI 1.72-4.73, p < 0.0001) and replacement (OR 2.33, 95% CI 1.55-3.55, p < 0.0001), no use of topical ice slush (OR 1.29, 95% CI 1.10-1.49, p < 0.0009), and use of inotropic agents for greater than 30 minutes after termination of cardiopulmonary bypass (OR 1.36, 95% CI 1.16-1.59, p < 0.0001). Postoperative median ICU stay (3.6 days AF vs. 2 days no AF, p < 0.001) and hospital stay (10 days AF vs. 7 days no AF, p < 0.001) were higher in AF. Morbid events, hospital mortality, and 6-month mortality were significantly higher in AF (p < 0.001): ICU readmission 13% AF vs. 3.9% no AF, perioperative myocardial infarction 7.41 % AF vs. 3.36% no AF, persistent congestive heart failure 4.57% AF vs. 1.4% no AF, reintubation 10.59% AF vs. 2.47% no AF, stroke 5.26% AF vs. 2.44% no AF, hospital mortality 5.95% AF vs. 2.95% no AF, 6-month mortality 9.36% AF vs. 4.17% no AF.

CONCLUSIONS

Atrial fibrillation after cardiac surgery occurs in approximately one third of patients and is associated with an increase in adverse events in all measurable outcomes of care and increases the use of hospital resources and, therefore, the cost of care. Strategies to reduce the incidence of AF after cardiac surgery should favorably affect surgical outcomes and reduce utilization of resources and thus lower cost of care.

Evaluations of retinopathy in the VA Cooperative Study on Glycemic Control and Complications in Type II Diabetes (VA CSDM). A feasibility study

OBJECTIVE

The main goal of the study of 153 male veterans was to determine whether a statistically and clinically significant difference in HbA1c could be achieved between a standard therapy and an intensively treated group of patients with type II diabetes. A second major goal was to assess the feasibility of collecting reliable high-quality endpoint data, including microvascular and macrovascular events. Retinopathy was defined as a key microvascular endpoint.

RESEARCH DESIGN AND METHODS

This was a randomized prospective trial of 153 men between the ages of 40 and 69 years, with type II diabetes for 15 years or less. Of the patients, 78 were assigned to the standard therapy arm and 75 to the intensive therapy arm. The goal of standard therapy was good general medical care and well-being and avoiding excessive hyperglycemia, glycosuria, ketonuria, or hypoglycemia. This was generally accomplished with one shot of insulin per day. The goal of intensive therapy was to obtain an HbA1c within two standard deviations of the mean of nondiabetic subjects (4.0-6.1%). This was obtained by a four-step management technique, with patients moving to the next step only if operational goals were not met. The steps were as follows: step 1: evening intermediate or long-acting insulin only; step 2: evening insulin with daytime glipizide; step 3: insulin, twice a day, no glipizide; and step 4: more than two injections of insulin, no glipizide. Retinopathy was assessed at baseline, 12, and 24 months by seven-field stereo fundus photography done at each of the five participating VA medical centers and read at the Central Reading Center at the Department of Ophthalmology, University of Wisconsin Medical School, Madison. Visual acuity was determined by ophthalmologists at each of the participating hospitals.

RESULTS

After the 6th month of the 24-month study, an average HbA1c of approximately 7.1% in the intensively treated group was sustained for the full study and was significantly lower than that seen in the standard group (9.2%, P < 0.001). Compliance in obtaining fundus photographs was excellent. Near normalization of glycemia did not cause transient worsening of retinal morphology nor did it prevent the onset or delay the progression of retinopathy. There was no effect on visual acuity.

CONCLUSIONS

1) A glycemic control intervention study in people with type II diabetes is feasible and safe; 2) intensive control did not cause transient deterioration of retinopathy; and 3) although no improvement was seen in retinopathy, the follow-up was 24 months, an interval shorter than the 3 years or more of intensive therapy before improvement is seen in type 1 diabetic studies. This does not rule out the possibility that longer periods of intensive therapy would have improved retinopathy. A full-scale intervention trial in type II diabetes is needed to resolve this issue.

Comparison of sulfasalazine and placebo in the treatment of ankylosing spondylitis. A Department of Veterans Affairs Cooperative Study

OBJECTIVE

To determine whether sulfasalazine (SSZ) at a dosage of 2,000 mg/day is effective for the treatment of active ankylosing spondylitis (AS) that is not controlled with nonsteroidal antiinflammatory drug therapy.

METHODS

Two hundred sixty-four patients with AS were recruited from 15 clinics, randomized (double-blind) to SSZ or placebo treatment, and followed up for 36 weeks. Treatment response was based on morning stiffness, back pain, and physician and patient global assessments.

RESULTS

While longitudinal analysis revealed a trend favoring SSZ in the middle of treatment, no difference was seen at the end of treatment. Response rates were 38.2% for SSZ and 36.1% for placebo (P = 0.73). The Westergren erythrocyte sedimentation rate declined more with SSZ treatment than with placebo (P < 0.0001). AS patients with associated peripheral arthritis showed improvement that favored SSZ (P = 0.02). Adverse reactions were fewer than expected and were mainly due to nonspecific gastrointestinal complaints.

CONCLUSION

SSZ at a dosage of 2,000 mg/day does not seem to be more effective than placebo in the treatment of AS patients with chronic, longstanding disease. SSZ is well tolerated and may be more effective than placebo in the treatment of AS patients with peripheral joint involvement. This effect is more pronounced in treatment of the peripheral arthritis in this subgroup of AS patients.