Risk of venous thromboembolism from oral contraceptives containing gestodene and desogestrel versus levonorgestrel: a meta-analysis and formal sensitivity analysis.

Controversy exists regarding whether oral contraceptives (OCs) containing desogestrel and gestodene are associated with an increased risk of venous thromboembolism (VTE) versus OCs containing levonorgestrel. We were interested in synthesizing the available data, exploring explanations for mixed results, and characterizing the degree of uncontrolled confounding that could have produced a spurious association. We performed a meta-analysis and formal sensitivity analysis of studies that examined the relative risk of VTE for desogestrel and gestodene versus levonorgestrel. Twelve studies, all observational, were included. The summary relative risk (95% CI) was 1.7 (1.3-2.1; heterogeneity p = 0.09). If real, the incremental risk of VTE would be about 11 per 100,000 women per year. An association was present when accounting for duration of use and when restricted to the first year of use in new users. However, in the sensitivity analysis, the association abated in many, but not all, scenarios in which an unmeasured confounding factor increased the risk of VTE three to fivefold and in nearly all examined scenarios in which the factor increased the risk 10-fold. The summary relative risk of 1.7 does not appear to be caused by depletion of susceptibles, but is sensitive to a modest degree of unmeasured confounding. Whether such confounding occurred is unknown. However, given this sensitivity, this issue probably cannot be settled unequivocally with observational data. In the absence of a definitive answer, this apparent increased risk, together with its uncertainty and small magnitude and its important consequences, should be considered when selecting an OC for a given woman.

Risk factors for infective endocarditis: oral hygiene and nondental exposures.

BACKGROUND: The risks of infective endocarditis (IE) associated with various conditions and procedures are poorly defined. METHODS AND RESULTS: This was a population-based case-control study conducted in 54 Philadelphia, Pa-area hospitals from 1988 to 1990. Community-acquired IE cases unassociated with intravenous drug use were compared with matched community residents. Subjects were interviewed for risk factors. Diagnoses were confirmed by expert review of medical record abstracts with risk factor data removed. Cases were more likely than controls to suffer from prior severe kidney disease (adjusted OR [95% CI]=16.9 [1.5 to 193], P:=0.02) and diabetes mellitus (adjusted OR [95% CI]=2.7 [1.4 to 5.2], P:=0.004). Cases infected with skin flora had received intravenous fluids more often (adjusted OR [95% CI]=6.7 [1.1 to 41], P:=0.04) and had more often had a previous skin infection (adjusted OR [95% CI]=3.5 [0.7 to 17], P:=0.11). No association was seen with pulmonary, gastrointestinal, cardiac, or genitourinary procedures or with surgery. Edentulous patients had a lower risk of IE from dental flora than patients who had teeth but did not floss. Daily flossing was associated with a borderline decreased IE risk. CONCLUSIONS: Within the limits of the available sample size, the data showed that IE patients differ from people without IE with regard to certain important risk factors but not regarding recent procedures.

Detailed examination of fenfluramine-phentermine users with valve abnormalities identified in Fargo, North Dakota.

Although several studies have reported on valve abnormalities among users of fenfluramine or dexfenfluramine, detailed information on these subjects has not been provided, limiting the ability to understand who may be at risk for valve abnormalities and to generate hypotheses about the etiology and pathogenesis of these abnormalities. This study was a detailed medical record review of 18 previously reported users of fenfluramine and phentermine, all with valve abnormalities on echocardiogram and 2 with surgical pathology. Both clinical characteristics and medication use were recorded by trained abstracters using a standardized data collection form. Two subjects (11%) had other possible etiologies of valve disease: a history of rheumatic fever and prescribed ergotamine. Three subjects (17%) had a history of migraine headaches and 4 (22%) had murmurs noted before using fenfluramine. Use of medications that may affect serotonin receptors was common: ergotamine (1 subject, 5%), selective serotonin reuptake inhibitors (6, 33%), sumatriptan (2, 11%), and mirtazapine (1, 5%). Prior medication and nonmedication allergies were recorded in 6 (33%) and 3 (17%) subjects, respectively. All subjects had symptoms possibly due to fenfluramine or phentermine side effects. This study raises the hypotheses that valvular heart disease among fenfluramine users may be less common than previously estimated, that serotonin excess may play a role in valve pathology, and that a patient's response to anorexigens and other medications may serve as a marker for increased risk. Further study is needed to test these hypotheses.

Predictors of injury among adult recreational in-line skaters: a multicity study.

OBJECTIVES: This study examined risk factors for injury, injury prevalence, safety gear use, and skating habits of adult recreational in-line skaters. METHODS: Randomly selected in-line skaters in 6 major US cities were interviewed. RESULTS: Only 6% of skaters consistently wore all 4 recommended types of safety gear. Skaters with greater skating experience were more likely to perform tricks, wear less safety gear, and sustain an injury. CONCLUSION: More experienced adult recreational in-line skaters are at increased risk for injury. Safety gear use in alarmingly low in adult recreational in-line skaters, especially experienced skaters. Safe skating education programs should consider targeting this newly recognized at-risk skating population.

Dental and cardiac risk factors for infective endocarditis. A population-based, case-control study.

BACKGROUND: Although antibiotic prophylaxis against infective endocarditis is recommended, the true risk factors for infective endocarditis are unclear. OBJECTIVE: To quantitate the risk for endocarditis from dental treatment and cardiac abnormalities. DESIGN: Population-based, case-control study. SETTING: 54 hospitals in the Philadelphia area. PATIENTS: Persons with community-acquired infective endocarditis not associated with intravenous drug use were compared with community residents, matched by age, sex, and neighborhood of residence. MEASUREMENTS: Information on demographic characteristics, host risk factors, and dental treatment was obtained from structured telephone interviews, dental records, and medical records. RESULTS: During the preceding 3 months, dental treatment was no more frequent among case-patients than controls (adjusted odds ratio, 0.8 [95% CI, 0.4 to 1.5]). Of 273 case-patients, 104 (38%) knew of previous cardiac lesions compared with 17 controls (6%) (adjusted odds ratio, 16.7 [CI, 7.4 to 37.4]). Case-patients more often had a history of mitral valve prolapse (adjusted odds ratio, 19.4 [CI, 6.4 to 58.4]), congenital heart disease (adjusted odds ratio, 6.7 [CI, 2.3 to 19.4]), cardiac valvular surgery (adjusted odds ratio, 74.6 [CI, 12.5 to 447]), rheumatic fever (adjusted odds ratio, 13.4 [CI, 4.5 to 39.5]), and heart murmur without other known cardiac abnormalities (adjusted odds ratio, 4.2 [CI, 2.0 to 8.9]). Among case-patients with known cardiac lesions--the target of prophylaxis--dental therapy was significantly (P = 0.03) less common than among controls (adjusted odds ratio, 0.2 [CI, 0.04 to 0.7] over 3 months). Few participants received prophylactic antibiotics. CONCLUSIONS: Dental treatment does not seem to be a risk factor for infective endocarditis, even in patients with valvular abnormalities, but cardiac valvular abnormalities are strong risk factors. Few cases of infective endocarditis would be preventable with antibiotic prophylaxis, even with 100% effectiveness assumed. Current policies for prophylaxis should be reconsidered.

Lack of hepatotoxic effects of parenteral ketorolac in the hospital setting.

BACKGROUND: No large controlled studies to date have examined the hepatic safety of parenteral ketorolac, which is used to treat acutely ill hospitalized patients who may be at greatest risk of liver injury. OBJECTIVE: To measure the association between the use of parenteral ketorolac and subsequent liver injury. METHODS: A nonexperimental cohort study conducted in 35 hospitals in the greater Philadelphia, Pa, region examined 10,272 courses of parenteral ketorolac (the exposed group) and 10,247 courses of parenteral opioid (the comparison group). Liver injury was defined by a modified international consensus definition that relied exclusively on liver function tests. Proportional hazards regression was used to calculate the rate ratio and 95% confidence interval for the association between ketorolac exposure and the occurrence of liver injury, controlling for potentially confounding factors, and to explore the possible effects of duration and dose. RESULTS: The incidence of liver injury was 1.0% in the ketorolac group and 1.2% in the opioid group, yielding an unadjusted rate ratio of 0.77 (95% confidence interval, 0.59 1.01). Simultaneously adjusting for multiple potentially confounding factors did not change this result. There was no evidence for a duration-response relationship (P = .96) or a dose-response relationship (P = .23). We were unable to identify any subgroups that were susceptible to possible hepatotoxic effects of parenteral ketorolac. CONCLUSIONS: This study failed to find evidence of a hepatotoxic effect of parenteral ketorolac use in the hospital setting and provides strong evidence against the existence of a clinically meaningful association between exposure to parenteral ketorolac in the hospital setting and liver injury.

An assessment of the usefulness of the Duke criteria for diagnosing active infective endocarditis.

We evaluated the usefulness of the Duke criteria for diagnosing cases of active infective endocarditis (IE). Patients were identified prospectively over a 3-year period at 54 hospitals in the Philadelphia metropolitan area. Three of us independently reviewed abstracted hospital records and classified 410 patients as definite, probable, or possible cases of IE or as probable noncases. We then applied the Duke criteria to this sample to assess the degree of agreement between our diagnoses and the diagnoses based on these new criteria. Agreement was good to excellent, ranging from 72% to 90%, depending on the case definition used. The sensitivity of the Duke criteria was also good to excellent, varying from 71% to 99%, again depending on case definition used. Specificity was lower (0-89%). We conclude that use of the Duke criteria will result in little underdiagnosis of IE but that it may result in overdiagnosis of IE; therefore, these criteria should be applied prospectively to determine their clinical usefulness.

Parenteral ketorolac: the risk for acute renal failure.

BACKGROUND: Acute renal failure has been associated with parenteral ketorolac tromethamine, but the risk that is associated with this therapy has not been quantified. OBJECTIVE: To compare the risk for acute renal failure associated with ketorolac with that associated with opioids. DESIGN: Retrospective cohort study. SETTING: 35 hospitals in or near Philadelphia. PATIENTS: Patients receiving 10,219 courses of parenteral ketorolac and patients receiving 10,145 courses of parenteral opioids. MEASUREMENTS: Acute renal failure was defined by 1) an increase in the serum creatinine concentration of 50% or more and 2) either an absolute increase of 44.2 mumol/L or more for concentrations that were less than 132.6 mumol/L at baseline or an absolute increase of 88.4 mumol/L or more for concentrations that were 132.6 mumol/L or more at baseline. In addition, a secondary definition required a diagnosis by a physician. RESULTS: The overall incidence of acute renal failure was 1.1% after therapy with either ketorolac or opioids. Multivariate-adjusted rate ratios comparing ketorolac with opioids for acute renal failure were 1.09 (95% CI, 0.83 to 1.42) overall, 1.00 (CI, 0.76 to 1.33) for less than 5 days of therapy, and 2.08 (CI, 1.08 to 4.00; P = 0.03) for more than 5 days of therapy. Similar results were obtained when the secondary definition of acute renal failure was used. CONCLUSIONS: Overall, acute renal failure was uncommon in this hospitalized population. Compared with opioids, ketorolac administered for 5 days or less did not increase the rate of renal failure. However, among patients who were treated with analgesics for more than 5 days, ketorolac may be associated with an elevated rate of acute renal failure.

Biochemical epidemiology of gallbladder cancer.

To evaluate the a priori hypotheses that an increased level of glyco and tauro lithocholic acid, perhaps because of a decreased capacity for hepatic sulfation, contributed to the biochemical epidemiology of gallbladder cancer, a case-control study was undertaken at four hospitals in La Paz, Bolivia, and at one hospital in Mexico City, Mexico. Eighty-four cases with newly diagnosed histologically confirmed gallbladder cancer were compared with 264 controls with cholelithiasis or choledocholithiasis in the absence of cancer and with 126 controls with normal biliary tracts. All study subjects were undergoing abdominal surgery. Interview data were collected for all study subjects, as well as blood, bile, and gallstone specimens when feasible. Sera were analyzed for carcinoembryonic antigen, cholesterol concentration, and total bile acids. Bile specimens were analyzed for carcinoembryonic antigen; and for concentration of bile salts; cholesterol; phospholipids; and the glycine and taurine conjugates of cholic, ursodeoxycholic, chenodeoxycholic, deoxycholic, and lithocholates; sulfoglycolithocholate; and sulfotaurolithocholate. Gallstone specimens were analyzed for the percentage of cholesterol content, the percentage of calcium bilirubinate content, and the percentage of calcium carbonate content. Serum bile acids were increased in cases versus the two control groups (median 11.7 nmol/mL vs. 9.3 nmol/mL for stone controls and 8.2 nmol/L for nonstone controls, P < or = .02 for each pairwise comparison). Biliary bile acids were markedly decreased in the cases (median 3.98 micromol/mL vs. 33.09 micromol/mL, and 154.0 micromol/L, respectively, P < or = .0001 for each comparison), even after excluding those with a serum bilirubin higher than 2.0 mg/dL. Bile cholesterol was lower for the cases as well (median 1.70 micromol/mL vs. 4.90 micromol/mL, and 16.81 micromol/ mL, respectively, P < or = .02), as was the concentration of bile phospholipids (median 2.97 micromol/mL vs. 6.26 micromol/mL, and 52.69 micromol/mL, P = .1 and .0004, respectively). Contrary to our a priori hypothesis, there was no difference between the cases and either control group in their bile concentrations of lithocholate, the proportion of bile acids which were sulfated, or the concentration of nonsulfated lithocholate. However, the cases had a higher concentration of ursodeoxycholate (UDC) (P < .004 for both control groups), especially glycoursodeoxycholate (P < .001 for both control groups). A previously published suggestion that gallstone size differed between cases and controls was not confirmed. In conclusion, cases with gallbladder cancer differed from controls with stones and from controls with normal biliary tracts in their serum and bile biochemistries. These findings may be a reflection of the disease process, or may provide useful clues to its pathogenesis.

Parenteral ketorolac and risk of gastrointestinal and operative site bleeding. A postmarketing surveillance study.

OBJECTIVE: To evaluate the risk of gastrointestinal and operative site bleeding associated with the use of parenteral ketorolac tromethamine. DESIGN: Postmarketing surveillance inception cohort study. SETTING: A total of 35 hospitals throughout the Philadelphia, Pa, region, 1991 to 1993. PATIENTS: Patients administered 10,272 courses of parenteral ketorolac therapy were compared with patients administered 10,247 courses of a parenteral opiate who were matched to the ketorolac patients by hospital, admitting service, and date of initiation of study drug. MAIN OUTCOME MEASURES: Medical records were reviewed for demographics, medical history, doses and duration of study drug, various aspects of the hospital course including surgery and concomitant medications, and adverse events. RESULTS: The multivariate adjusted odds ratio (OR) comparing ketorolac with opiates for gastrointestinal bleeding was 1.30 (95% confidence interval [CI], 1.11 to 1.52); for operative site bleeding, the OR was 1.02 (95% CI, 0.95 to 1.10). The OR was elevated further in subjects 75 years of age or older for both gastrointestinal bleeding (OR = 1.66; 95% CI, 1.23 to 2.25) and operative site bleeding (OR = 1.12; 95% CI, 0.94 to 1.35). A dose-response relationship was evident between average daily ketorolac dose and both gastrointestinal bleeding and operative site bleeding (trend test P < .001 for both). When analgesic therapy lasted 5 or fewer days, ketorolac was associated with only a small increased risk of gastrointestinal bleeding (OR = 1.17; 95% CI, 0.99 to 1.30); when therapy was prolonged beyond 5 days, the OR was 2.20 (95% CI, 1.36 to 3.57). The association of ketorolac with operative site bleeding was not affected by duration of therapy. CONCLUSIONS: The overall associations between ketorolac use and both gastrointestinal bleeding and operative site bleeding are small. However, the risk associated with the drug is larger and clinically important when ketorolac is used in higher doses, in older subjects, and for more than 5 days. Improving physicians' prescribing practices by limiting the dose and duration of ketorolac use, especially in the elderly, should enhance its overall risk-benefit balance.

Risk factors for gallbladder cancer. An international collaborative case-control study.

BACKGROUND: Gallbladder cancer has an unusual geographic and demographic distribution, suggesting many possible etiologies. METHODS: A case-control study was undertaken at four hospitals in La Paz, Bolivia, and at one hospital in Mexico City, Mexico. Eighty-four patients with newly diagnosed, histologically confirmed gallbladder cancer were compared with 126 control subjects without stones and with 264 control subjects with cholelithiasis or choledocholithiasis without cancer. All study subjects underwent abdominal surgery. Study subjects were interviewed regarding demographic characteristics, medical history, family history, diet, and exposure to agents presumed to be risk factors for biliary cancer. RESULTS: Virtually all subjects in Mexico were judged to be mestizos (i.e., persons of mixed ancestry) In contrast, race was a very strong risk factor for gallbladder cancer in Bolivia. Relative to mestizos who spoke neither language, the odds ratio (95% confidence interval [CI]) for cases versus control subjects without stones for those who spoke Aymara well was 15.9 (CI, 1.9-179), whereas it was 1.4 (CI, 0.2-8.2) for those who spoke Quechua well. An increased risk was also noted for elevated maximum body mass index (P = 0.03), family history of gallstones (odds ratio [OR] = 3.6 [CI, 1.3-11.4]), and physician-diagnosed typhoid (OR = 12.7 [CI, 1.5-598]). An increased risk was also seen with elevated maximum body mass index; compared with those with a body mass index less than 24 kg/m2, those with an index of 24-25 kg/m2, 26-28 kg/m2, and greater than 28 kg/m2 had odds ratios of 1.6 (CI, 0.4-7.6), 1.3 (CI, 0.3-5.6), and 2.6 (CI, 0.5-18.6), respectively (asymptotic test for trend, P = 0.03). Finally, a number of associations were noted with certain dietary and cooking habits. CONCLUSIONS: Patients with gallbladder cancer differed from control subjects in race, body mass, physician-diagnosed typhoid, and certain dietary patterns. These findings may provide useful clues to the pathogenesis of gallbladder cancer, but given the number of analyses performed, additional cases need to be studied.

Incidence of infective endocarditis in the Delaware Valley, 1988-1990.

This population-based study aimed to determine the incidence of native, prosthetic, and bioprosthetic valve nosocomial infective endocarditis (IE), and IE associated with the use of injected drugs. Patients with IE during 27 months over the years 1988 to 1990, and residing in any of 6 counties in the Philadelphia metropolitan area were identified. An expert panel reviewed all patients to verify the diagnosis. Incidence rates were estimated after adjustment for failure to recruit and underreporting. Of 853 potential patients, 670 (79%) met the inclusion criteria. The overall incidence rate of IE was 11.6 cases/100,000 person-years (95% confidence interval [CI] 10.8 to 12.4). The rates for specific types of IE were: 4.45 (95% CI 3.97 to 4.94) for community-acquired native valve, 0.94 (95% CI 0.72 to 1.12) for prosthetic valve, 0.94 (95% CI 0.71 to 1.16) for nosocomial, and 5.34 (95% CI 4.80 to 5.87) for IE associated with use of injected drugs. Previous population studies found overall incidence rates of 1.7 to 4 cases/100,000 person-years, similar to our rate for community-acquired native valve IE. Type-specific rates have not been previously reported. The higher overall rate in this study is partly related to the high prevalence of injection drug use in our area.

Assessing diagnostic criteria for active infective endocarditis.

A study was performed to describe agreement among experts on their classification of patients, in the absence of information concerning risk factors, as to the presence of infective endocarditis (IE). The study also assessed the clinical characteristics that enabled the experts to determine that a patient had IE. All patients with a discharge diagnosis of IE were identified prospectively from 54 hospitals in the Delaware Valley over a 3-year period. Patients were part of a case-control study of risk factors for IE. Three infectious disease experts independently reviewed abstracted hospital records and classified each of 151 eligible patients as a definite, probable or possible case, or a probable noncase, both based on clinical judgement and using a modified standard definition. Experts were more likely to classify a patient as a definite case of IE on the basis of clinical judgement than by using the modified standard definition. Agreement between reviewers was 92 to 95% when they were distinguishing only probable non-cases from others. Agreement between reviewers on specific categories was lower (40 to 58%). The number of positive blood cultures was a strong predictor of a patient's being classified as a case, as was the type of infecting organism. It is concluded that experts are willing to make a definitive diagnosis of IE on the basis of blood culture information alone. Further supporting evidence, such as the presence of vegetation on an echocardiogram, is needed when blood culture results are ambiguous.

The utility of preoperative electrocardiograms in the ambulatory surgical patient.

To determine the utility of preoperative screening electrocardiograms (ECGs) among ambulatory surgery patients, we reviewed the charts of 751 consecutive adult patients who underwent ambulatory surgery. Data were collected on demographic characteristics, coexisting medical problems, American Society of Anesthesiologists physical status score, preoperative ECG results, adverse intraoperative cardiovascular events, and postoperative cardiovascular complications. In our study population of relatively healthy outpatients, preoperative ECGs were abnormal in 42.7% of patients. Age, increased physical status score, and male gender were associated with a greater incidence of abnormal preoperative ECGs. There were 12 adverse cardiovascular perioperative events among the 751 patients (1.6%), and the preoperative ECG may have been clinically useful in six of these 12 patients. Neither preoperative ECGs nor results of preoperative screening questionnaires were predictive of adverse cardiovascular perioperative events. These findings question the utility of preoperative ECGs in the ambulatory surgery setting, especially among younger, relatively healthy patients.

Routine use of the prothrombin and partial thromboplastin times.

The prothrombin time (PT) and activated partial thromboplastin time (APTT) tests are often routinely ordered for hospitalized patients. Ordering patterns and clinical indications for the PT and APTT tests on the medical service at a teaching hospital were studied. Eighty-one percent of all patients admitted to the medical service had a PT and APTT test ordered. When compared with a modified version of guidelines for the use of the PT and APTT tests recently developed by the Medical Necessity Project of the Blue Cross and Blue Shield Associations of America and endorsed by the American College of Physicians, at least 70% of these tests were not clinically indicated. Many of the unindicated tests were ordered prior to invasive procedures or, apparently, out of habit. These inappropriate PT and APTT tests cost at least $60,948 per year for the medical service. Based on these findings, we suggest methods of reducing the inappropriate use of the PT and APTT tests.

Cancer patients' knowledge, beliefs, and behavior regarding pain control regimens: implications for education programs.

The inadequate control of cancer-related pain is a matter of great concern. Misconceptions about pain medications on the part of professionals and the public have been implicated as causes of this problem. Another cause may be noncompliance with prescribed regimens. A study of 82 cancer patients who were prescribed pain medications was conducted to further investigate the problem of pain control in this population. Patients were interviewed twice: on the day after they received pain prescriptions (T1) and again two weeks later (T2). There was a significant decrease in patients' reported pain between T1 and T2 (P = 0.0001). By their own report, the level of patients' compliance with prescribed pain control regimens was very high. At T2 77% of the patients recalled the correct names of their pain medications and 88% had been taking them. However, an important gap in patients' knowledge was revealed by their general inability to recall any of the common side effects of pain medications (constipation, nausea, and sedation). None of the measures of pain--pain level at T1 and T2 and change in pain between T1 and T2--was related to the patients' education, age, attending physician, prescribed medication schedule, or medication strength. More women than men reported increased pain during the study period (P = 0.04). In general, the patients' concern about possible addiction or tolerance to prescribed medications were low. However, there was a positive relationship between intensity of pain and concern about tolerance to drugs (P = 0.0003).(ABSTRACT TRUNCATED AT 250 WORDS)