Robert H. Noth

The effects of alcohol on the endocrine system.

Male hypogonadism is the best documented endocrine effect of chronic alcoholism. A reversible clinical syndrome resembling Cushings syndrome has also recently been described in some chronic alcoholics. The pituitary-thyroid axis is relatively resistant to the effects of ethanol, although mild abnormalities in various thyroid tests are frequently noted in the presence of alcoholic liver disease.

The effects of alcohol on blood pressure and electrolytes.

The interaction between alcohol abuse, changes in blood pressure, and electrolyte abnormalities is complex. Some effects of alcohol are seen only with acute ingestion, some during withdrawal, and some only in chronic drinkers. Careful attention to the interactions between the metabolism of various electrolytes can prevent unnecessary morbidity and mortality in alcoholic patients.

Computer-based screening of veterans for metabolic syndrome.

BACKGROUND The worldwide obesity epidemic has focused attention on the prevalence of the metabolic syndrome, which is greater than previously believed. However, estimates of its overall prevalence vary widely. We wished to ascertain the prevalence of metabolic syndrome in the Veteran Affairs Northern California Health Care System (VANCHCS). METHODS We used the VA computerized clinical database and modified Adult Treatment Panel III (ATP III) criteria of fasting blood glucose (FBG) >or=110 mg/dL, blood pressure >or=130/>or=85, triglyceride >or=150 mg/dL, high-density lipoprotein cholesterol (HDL-C) <40 mg/dL (men) or <50 mg/dL (women), but body mass index (BMI) >or=30 in lieu of waist circumference >102 cm (40 inches) for men and 88 cm (35 inches) for women. We also accepted current pharmacotherapy for diabetes as qualifying for elevated fasting blood sugar (FBS); current therapy with niacin, gemfibrozil, or fenofibrate for elevated triglyceride concentrations; and recent use of multiple International Classification of Diseases, 9(th) Revision (ICD-9) codes for hypertension for elevated blood pressure. RESULTS We examined all clinical records for veterans registered in VANCHCS who filled any prescription between July 1, 2004, and June 30, 2005 (n = 51,026). Their average age was 63 years; 93% were male. In all 25% (n = 13,010) were diagnosed as having metabolic syndrome by meeting at least 3 of the above 5 criteria. Because only 60% (n = 30,727) of the population had data for 3 or more criteria, the actual percent with metabolic syndrome is probably substantially higher. CONCLUSIONS Over one quarter of veterans in the VANCHCS may have metabolic syndrome based on our modified ATP III criteria. We urge screening more veterans with fasting laboratory testing. Computerized screening of a large clinical database can provide an effective strategy to aid clinicians in identifying more patients at risk for cardiovascular disease.

Blood pressure control in a diabetic population assessed by computer review

Recent data from multicentre randomized controlled clinical trials suggest that tight BP management has greater benefit than strict glucose control in patients with diabetes. 1,2 In recognition of this, the American Diabetes Association (echoing recommendations from the sixth and seventh Joint National Councils for Hypertension (JNC 6 and VII 3 ) recommends a goal BP < 130/85 mm Hg for patients with diabetes.

Long-Term Maintenance of Glucose Control in Veterans with Type 2 Diabetes Mellitus Using Oral Agents

PURPOSE OF STUDY The aim of this study was to evaluate long-term glycemic control in individuals with type 2 diabetes mellitus on oral hypoglycemic agents. METHODS We identified the cohort of veterans prescribed hypoglycemic agents every year from July, 1992, through June, 2007 (n=191). Glycosylated hemoglobin (HbA1c) was used to assess glycemic control. Data are expressed as mean±standard deviation (SD); statistics are expressed by t-test and chi-squared. P<0.05 was considered significant. RESULTS In the first year, 96 of the select group of 191 veterans identified above received oral agents only (OAO), 74 insulin only, and 21 both insulin and oral agents. Fifteen years later, 59 were OAO, 78 insulin only, and 54 both. Six patients receiving insulin in 1992-1993 were OAO-treated in 2006-2007. In the subgroup on OAO both at baseline and at the end (n=53), HbA1c decreased from 7.89±1.21 to 7.09±1.13 (P<0.001). These veterans were older at baseline (62.4±6.2) and leaner at the 15-year follow-up [body mass index (BMI) 28.1±4.9] than those who received insulin in 2006-2007 (n=43; age=57.9±9.6; BMI=32.3±7.9; P<0.05 and 0.005, respectively). Patients in the stable OAO group (n=53) were 74.0% Caucasian, compared to 51.2% in former-OAO [n=43; P<0.05 (chi-squared)]. CONCLUSIONS Over half (n=53; 55%) of patients originally in the OAO group remained so 15 years later. These stable patients were in better glycemic control, both at baseline and follow-up, less obese, older, and more likely to be Caucasian, than those who eventually received insulin. Currently used oral agents often maintain, or even improve, glucose control, over 2 decades after diagnosis of diabetes mellitus.

Computer confirmation of improved blood pressure control in diabetic patients.

In order to evaluate efforts to improve blood pressure (BP) control in veterans with diabetes, the authors performed a retrospective review of computerized health records in the Veterans Affairs Northern California Health Care System (VANCHCS). A total of 6624 diabetic patients with BP data were identified and a comparison was made to earlier surveys. There was an overall improvement in BP control. In a cohort of 1791 patients followed since 1999, there was a progressive improvement in both systolic BP (patients at goal increased from 28% to 35%; P<.001, chi(2)) and diastolic BP (patients at goal increased from 69% to 79%; P<.001). There was a corresponding increase in both number of drug classes used and use of antiangiotensin treatment. The authors conclude that the improvement in BP control in these veterans with diabetes was temporally associated with intensification in antihypertensive pharmacotherapy and sustained provider education and patient-specific computerized reminders.

TEN-YEAR PERSPECTIVE ON THERAPY FOR ERECTILE DYSFUNCTION IN MEN WITH DIABETES: VASTLY IMPROVED BUT STILL NOT FOR ALL.: 96

Based on reports in the literature, the estimated prevalence of erectile dysfunction (ED) is over 50% in men with diabetes. In the VA Northern California Health Care System (VANCHCS) in 1996, prior to use of alprostadil and sildenafil, only 2% of our male veterans with diabetes (MVDM) were receiving any therapy (Rx) for ED. By 2001, this figure had increased, but only to 9%, mostly due to use of sildenafil (begun in 2000). With alprostadil and phosphodiesterase 5 (PDE-5) inhibitors on formulary in VANCHCS for 6+ years, free or at minimal cost (

127 COMPUTER-BASED SCREENING OF VETERANS FOR METABOLIC SYNDROME.

7/4doses/month) and widely promoted in the media, prescriptions for these are likely to reflect current demand for Rx of ED. To assess this in our system in the past year (7/1/2005-6/30/2006) and back to 1996, VA computer-based records were examined for prescriptions for all pharmacologic Rx for ED and for orders for vacuum devices and ICD-9 codes for penile implants. MVDM were identified by use of hypoglycemic agents (7/1/2005-6/30/2006). The study group was composed of 8,894 MVDM and 39,827 male veterans without diabetes (MVnoDM) who received at least one prescription of any kind in the time interval. Fewer than 1% received nonpharmacologic ED Rx and were excluded from the analysis; 1,388 (16%) of MVDM received at least one prescription for a PDE-5 and 1,503 (17%) received any ED agent in the past year, almost double the treatment rate in 2001. An additional 1,102 (42% of those ever treated) received treatment in the past but dropped out before this past year. MVDM treated for ED were younger (62 ± 8 vs 67 ± 11 years; p

289 COMPUTER-BASED SCREENING FOR COMPRESSION FRACTURES AS AN INDEX OF OSTEOPOROSIS

In part as a result of an obesity epidemic, the prevalence of the “metabolic syndrome” is greater than previously believed. However, despite growing public and professional attention to the condition, estimates of its overall prevalence vary widely. Because of the potential impact of this syndrome on VA health care, we wished to ascertain the prevalence in the VANCHCS using the VA computerized clinical database. To do so, we modified the ATPIII criteria, keeping fasting blood glucose (FBS), blood pressure, triglyceride, and HDL cholesterol criteria but substituting BMI ≥ 30 for waist:hip ratio. We also accepted current pharmacotherapy for diabetes as qualifying for elevated FBS; current therapy with niacin, gemfibrozil, or fenofibrate for elevated triglyceride; and recent use of multiple ICD-9 codes for hypertension for elevated blood pressure. We examined all clinical records between July 1, 2004 and June 30, 2005 for veterans registered in VANCHCS who filled any prescription during this interval (n = 51,026). Their average age was 63 years; 93% were male. Twenty-five percent (n = 13,010) were diagnosed as having metabolic syndrome by meeting at least three of the above 5 criteria. Since only 60% (n = 30,727) of the population had data for 3 or more criteria, the actual percent with metabolic syndrome is probably substantially higher. We conclude that over one-quarter of our veterans in the VANCHCS may have metabolic syndrome based on our modified ATPIII criteria. We need to be screening more of our veterans with fasting laboratory testing. This computerized screening of a large clinical database can then provide an effective strategy to aid clinicians in identifying more patients at risk for cardiovascular disease.

142 COMPUTER-BASED ASSESSMENT OF GLUCOSE CONTROL IN THE VA NORTHERN CALIFORNIA HEALTH CARE SYSTEM: THE UNFINISHED BUSINESS

We, and others, have reported that osteoporosis in men is more prevalent than previously believed. However, despite growing attention to the condition, screening strategies driven by unreliable historical data such as height loss, are relatively insensitive and non-specific. Because of our concern that many men with osteoporosis may not be identified as at-risk, we performed a computer-based review of chest x-ray (CXR) reports in the VA VISTA clinical database, looking for unrecognized vertebral fractures. We examined all CXR reports between July 1, 2001 and June 30, 2003, scanning for the terms “compression” or “wedg- (to encompass “wedge” or “wedging”)” but without the negative modifiers “no compression” or “no acute compression” or “no fracture.” During the specified time period, 26,994 CXR were performed on 18,069 patients. 22,494 (83.3% of the total) CXR were done in men