Bone infections involving anaerobic bacteria.

Over 700 cases of anaerobic osteomyelitis have been reported in the literature. Nonetheless, most reviews of osteomyelitis have paid little attention to the potential role of anaerobes in bone infections. There have, as yet, been no prospective studies of osteomyelitis utlizing optimal anaerobic transport and culture techniques. In a retrospective study of osteomyelitis at Wadsworth VA Hospital from 1973--1975, 39 percent of 58 patients with osteomyelitis had an infection involving anaerobes. Anaerobes were isolated from 81 percent of 27 patients whose specimens were cultured anaerobically. Anaerobes were isolated from nine of ten samples of bone. Anaerobic bacteria were part of a mixed flora involving facultative bacteria in all but two cases. All of the patients with anaerobic infection had non-hematogenous osteomyelitis. Non-hematogenous disease comprises 80--90 percent of the osteomyelitis seen in adults. Our experience at Wadsworth VA Hospital and a review of the literature lead us to believe that anaerobes play a much larger role in osteomyelitis than has been appreciated previously. Infections of the calvarium, mastoid, mandible, maxilla and the extremities are most likely to involve anaerobes. Predisposing conditions include paranasal sinusitis, otitis media, periodontal disease, trauma, peripheral vascular disease, peripheral neuropathy and/or chronic osteomyelitis. The presence of a foul odor is a valuable clinical clue to the presence of anaerobes. Bacteroides, fusobacteria and anaerobic cocci have been reported with almost equal frequency from anaerobic bone infections. While Bacteroides fragilis is the most common anaerobe isolated in infections of other organ systems, it does not appear to be a common pathogen in anaerobic bone infections. The role of anaerobes in osteomyelitis is not yet resolved. They have been isolated in pure culture from infected bone, and under those circumstances are clearly pathogenic. Anaerobes are found more frequently as part of a mixed flora with facultative streptococci, gram-negative bacilli, and less often with S. aureus. In this setting it is unclear which organism or organisms are the primary invaders, or whether there is a synergistic mechanism of infection. The reliability of sinus drainage cultures also remains to be determined. Our retrospective study suggests that certain anaerobes isolated from sinus drainage are not present in infected bone. Cultures of bone or an abscess adjacent to bone would be expected to give more reliable data. The majority of anaerobes other than B. fragilis are susceptible to levels of penicillin achievable with parenteral administration of the antibiotic. Anaerobic pathogens should be sought in the situations noted above. We feel that parenteral penicillin should be part of the initial antibiotic regimen in patients with suspected or documented anaerobic bone infection...

Amikacin therapy for gram-negative septicemia.

Amikacin was administered to 18 patients with gram-negative septicemia. Ten of the patients had blood culture isolates highly resistant to gentamicin; six of these patients had persistent bacteremia while receiving gentamicin alone or in combination with other agents. Fourteen of the 18 patients were cured with amikacin therapy and adjunctive measures. Nine of the 10 patients with gentamicin-resistant pathogens were cured. The occurrence of nephrotoxicity in four patients with elevated amikacin serum levels and serious underlying disease indicates the desirability of monitoring serum amikacin levels. Minor ototoxicity occurred in two patients and was associated with prolonged therapy and high serum amikacin levels. Amikacin is a highly effective agent for treating patients with gram-negative bacteremia; it is the agent of choice in the therapy of patients with suspected or documented gram-negative bacteremia caused by pathogens resistant to gentamicin and susceptible to amikacin.

Cardiac function in Duchenne's muscular dystrophy. Results of 10-year follow-up study and noninvasive tests.

The purposes of this study were to: (1) evaluate the progression of cardiac involvement in Duchenne's muscular dystrophy using systolic time intervals (PEP/LVET); (2) determine if the degree of cardiac involvement bears a relation to the severity of skeletal muscle disease; and (3) describe the M-mode and two-dimensional echocardiographic findings. In 1970, systolic time intervals were studied in 16 patients. During the 10-year interim, two patients were lost to follow-up study, and five patients died. Nine remaining patients were re-studied in 1980 with M-mode and two-dimensional echocardiography as well as systolic time intervals. The PEP/LVET value of these nine patients increased from 0.37 +/- 0.05 (mean +/- SD) in 1970 to 0.47 +/- 0.07 (p less than 0.005) in 1980. Three patients remained ambulatory, and their PEP/LVET value (0.41 +/- 0.04) was significantly better than that of the nonambulatory patients (0.50 +/- 0.07, p less than 0.05). The M-mode echocardiography percentage diameter change was also worse in the nonambulatory group (21 +/- 4 percent versus 34 +/- 7 percent, p less than 0.02). The five patients who were nonambulatory in 1970 died in the intervening 10 years. This study demonstrated that the heart disease of Duchenne's muscular dystrophy is progressive and that the severity of skeletal muscle disease is probably associated with the degree of cardiac dysfunction.

Clinical use of serum digoxin concentrations.

The development of the radioimmunoassay for digoxin by Smith and coworkers in 1969 was a landmark in digitalis therapy. Since then, the complex pharmacokinetics of digoxin have been defined. As a result, the incidence of digitalis toxicity has markedly decreased. To use the digoxin assay properly, however, the relation of this pharmacokinetic parameter to digoxin pharmacodynamics must be known and the limitations of the assay itself understood. Systolic time intervals (STI) are uniquely useful to quantitate the inotropic effect of digitalis preparations. This technique can demonstrate the onset and magnitude of the inotropic effect for both oral and intravenous digitalis administration. By defining the mathematical relation between STI and simultaneous serum digoxin concentrations following intravenous administration of 1 mg digoxin, computer simulations can be made of the effect of dosing changes on blood and tissue concentrations. The serum digoxin assay has technical problems relating to quality control, interference by metabolites, and cross-reactions with endogenous digitalis-like substances. Further, a standard time for measurement following dosing has not been established. Physical activity can significantly after the serum digoxin concentrations by increasing skeletal muscle binding. Numerous drugs can interfere with digoxin absorption or elimination. Using the serum digoxin assay is the only way to assess these interactions. Computer surveillance (ideally with physician or pharmacist interaction) has been used to monitor digitalis but has not yet gained widespread acceptance. This is clearly a method in need of further testing.

Beneficial effects of atropine in the pre-hospital phase of coronary care.

The current widespread interest in early (pre-hospital) coronary care has raised questions regarding the safety and efficacy of the use of atropine in these circumstances. Although reports from early coronary care systems have been favorable, serious questions of safety have been raised on the basis of experimental animal studies and isolated case reports. A sample of the experience of a large scale emergency coronary care system in Columbus, Ohio has therefore been reviewed. A group of 70 patients who, during the early phases of myocardial infarction, exhibited bradyarrhythmia (less than 60 beats/min) was identified. When this group was further categorized on the basis of arterial pressure at the time of initial observation, patients with normal or elevated blood pressure had low mortality rates whether treated with atropine or not. In patients with hypotension the mortality rate was 75 percent without therapy, and 25 percent with therapy. Indeed, when normal hemodynamics were restored, the mortality rate was only 11 percent. Unexpected ventricular fibrillation occurred in 1 of 45 treated patients and 2 of 25 untreated patients. Although the study was not controlled, the data strongly suggest that atropine is both beneficial and safe for use in treating bradyarrhythmia and hypotension in the early phase of acute myocardial infarction.

Usefulness of systolic time intervals in coronary artery disease.

This review summarizes current knowledge concerning the value of systolic time intervals in coronary artery disease. Although the usual pattern of prolongation of the preejection period (PEP) and shortening of the left ventricular ejection time (LVET) characteristic of left ventricular failure is seen in acute myocardial infarction, the systolic time intervals (as well as all other measures) are profoundly influenced by adrenergic hyperactivity characteristics of this disorder. Adrenergic stimulation normally shortens both the PEP and LVET indexes and decreases the PEP/LVET ratio. The degree of shortening of electromechanical systole (QS2) is directed related to the excessive adrenergic tone. Patients with the greatest systolic time interval abnormalities have a poorer prognosis, a greater incidence of congestive heart failure and more abnormalities of directly measured indexes of left ventricular performance. The systolic time intervals are useful for assessing left ventricular performance in chronic coronary artery disease as well. In chronic coronary artery disease the PEP/LVET ratio and angiographically determined left ventricular ejection fraction are closely correlated ( r = -0.76), but the level of this correlation is less than that in other forms of left ventricular disease. The left ventricular ejection time index is prolonged after exercise in patients with angina pectoris when compared with findings in normal subjects. Failure of the ischemic ventricle to respond to adrenergic stimulation is the most likely mechanism. Addition of the postexercise left ventricular ejection time to standard treadmill stress testing identifies a significant number of patients (23 percent) who would have had false negative results by electrocardiographic criteria alone. In addition, this index provides confirmatory evidence in those with apparently positive electrocardiographic test data. The systolic time intervals have been useful in assessing both medical and surgical therapy in coronary artery disease. The test can be performed repeatedly and provides a measure of both left ventricular performance and extent of adrenergic hyperactivity. Thus, evaluation of therapy represents the most useful future application of systolic time intervals.

Enhanced diagnostic power of exercise testing for myocardial ischemia by addition of postexercise left ventricular ejection time.

To improve both the sensitivity and specificity of the multistage treadmill test, postexercise systolic time intervals were prospectively studied in 73 patients with angina-like chest pain and normal resting ST-T segments. The decision to perform coronary angiography was made independent of the exercise test. Twenty-three patients had normal coronary arteries and 50 had more than 50 percent reduction of luminal diameter of one or more major coronary arteries. The systolic time intervals were measured before and 2,4,6,8 and 10 minutes after exercise. Of all the systolic time intervals, the 4 minute postexercise left ventricular ejection time proved most discriminating between normal subjects and those with coronary artery disease. The deviations of this interval from the normal regression with heart rate both before and after exercise were used to calculate the net delta left ventricular ejection time after exercise. A net increase of more than 31 msec represented 2 standard deviations above normal. Twenty-three (46 percent) of the patients with coronary artery disease had an abnormal net delta ejection time after exercise. Twenty-five (50 percent) had a positive electrocardiographic response with a 9 percent false positive rate. Thirteen (26 percent) had only a positive postexercise net delta ejection time so that a total of 76 percent of patients with coronary artery disease were identified. Thus, measurement of the postexercise net delta ejection time a simple and useful adjunct to multistage treadmill testing.

Effect of propranolol on postexercise left ventricular ejection time index.

The net delta left ventricular ejection time index 4 minutes after exercise is prolonged in many patients with coronary artery disease. This prolongation is thought to be due to the lack of response of the ischemic myocardium to adrenergic stimulation and has been proposed as a measure of myocardial ischemia. In this study, the effect of beta adrenergic blockade on net delta left ventricular ejection time was studied in nine normal subjects (Group A) and in eight patients with stable angina and coronary artery disease (Group B). In Group A, a treadmill exercise test was performed for 10 minutes before and after administration of propranolol, 160 mg daily, for 2 days. The postexercise net delta left ventricular ejection time was significantly greater after propranolol (mean +/- standard error of the mean 12 +/- 4 versus 35 +/- 4 ms, p less than 0.01). In group B a maximal treadmill exercise test was performed before and after therapy with propranolol. Only patients with a normal net delta left ventricular ejection time before propranolol were selected. The net delta left ventricular ejection time again increased significantly after propranolol (11.5 +/- 4 versus 35.3 +/- 5 ms p less than 0.01). It is concluded that prolongation of postexercise net delta left ventricular ejection time cannot be used to diagnose ischemia in patients who are receiving propranolol therapy. Our data support the hypothesis that prolongation of net delta left ventricular ejection time after exercise is caused by an impaired myocardial response to catecholamines, whether due to ischemia or effective beta adrenergic blockade.

Prolonged electrical systole and QT greater than QS2 secondary to coronary artery disease.

Dyssynchronous depolarization-repolarization in the left ventricular (LV) myocardium may produce QT greater than QS2 or long QT. In 41 patients with coronary artery disease (CAD) and LV aneurysm, 46 patients with CAD and a history of acute myocardial infarction (AMI) but no LV aneurysm, and 52 patients with CAD without previous AMI, QT and QS2 were measured simultaneously at a speed of 100 mm/s within 48 hours of cardiac catheterization. Patients receiving class I antiarrhythmic drugs were excluded. The incidence of QT greater than QS2 was significantly greater in patients with LV aneurysm (71%) than in those with previous AMI (22%) and those with CAD but no previous AMI (20%) (p less than 0.001). Likewise, the incidence of long QT corrected for heart rate was significantly greater in patients with LV aneurysm (54%) than in those with previous AMI (7%) and those with CAD and no previous AMI (6%) (p less than 0.0001). The incidence of QT greater than QS2 in another 19 patients with previous AMI who were receiving digitalis therapy was significantly greater (65%) than in those with previous AMI but not receiving digitalis therapy (22%) (p less than 0.001). The incidence of long QT corrected for heart rate and QT greater than QS2 was not statistically different between patients with previous AMI and those with CAD but no previous AMI. The QT greater than QS2 or long QT in patients with aneurysm is probably a result of dyssynchronous depolarization or repolarization within or in the border zone of the LV aneurysm.(ABSTRACT TRUNCATED AT 250 WORDS)

Positive inotropic effects of hydralazine in human subjects: comparison with prazosin in the setting of congestive heart failure.

Hydralazine possesses positive inotropic properties in animal model preparations. To determine whether this vasodilator elicits a positive inotropic response in the failing human ventricle, hydralazine, 75 or 100 mg, was administered orally to 14 patients with left ventricular dysfunction and congestive heart failure; the results were compared with those of first-dose prazosin at 5 and 10 mg. The duration of the preejection period and the isovolumic developed pressure/duration of isovolumic contraction (delta P/delta t) were used as indexes of inotropy. Prazosin did not effect a significant response in the duration of the preejection period or in the delta P/delta t. Hydralazine significantly shortened the preejection period and increased the delta P/delta t over 8 hours after administration; these data suggest that hydralazine elicits a positive inotropic response in the failing human ventricle.

Syncope.

Syncope is a clinical entity of diverse cause. The historical features surrounding the syncopal event and the presence or absence of heart disease are the most important features in establishing the cause for syncope. Passive head-up tilt study provides a means of identifying many patients with vasodepressor syncope. Electrophysiologic study is important in the elucidation of syncope in patients who have syncope undefined after noninvasive evaluation. With proper use of the modalities available, few patients will have an undefined cause for syncope.

Linear relationship between electrical systole, mechanical systole, and heart rate.

The relationship between the duration of electrical systole (QT) and heart rate (HR) and the relationship between the QT interval and total electromechanical systole (QS2) were studied in the resting state in 200 patients (100 males and 100 females) without evidence of cardiovascular disease. A linear relationship was found between the QT and HR in males and females (males, QT = 521 msec - 2.0 HR, r = .91; females, QT = 511 msec - 1.8 HR, r = .90). In 20 male and 20 female subjects, the relationship between QT and QS2 was studied. The QT was slightly shorter but paralleled the QS2 (males QT = 529 msec - 2.1 HR, QS2 = 541 msec - 2.2 HR; females QT = 511 msec - 1.9 HR, QS2 = 540 msec - 2.0 HR). Thus, over the physiologic range of resting HR, a linear relationship exists between QT and HR. The QT interval is slightly shorter but parallels the QS2 in patients without heart disease. These linear relationships permit a direct comparison of the duration of electrical and mechanical systole.

Extensive myocardial blood flow distribution through individual coronary artery bypass grafts.

The regional myocardial perfusion distribution of coronary artery bypass grafts were studied in 61 patients who received 162 grafts. Selective intragraft instillations of radioactive-labeled macroaggregated albumin particles were used to study perfusion. The extent of individual graft perfusion was assessed in 100 patent grafts. Regional myocardial blood flow distribution was similar to the blood flow distribution of the native vessel receiving the graft in 64 of 100 grafts and less than that expected of the native vessel in 12 grafts. However, 24 grafts demonstrated a blood flow distribution pattern which extended beyond the normal distribution expected of the native vessel receiving the graft. This extensive perfusion could be attributed to collateral vessels or retrograde flow. A high incidence of both graft and native vessel occlusion was found in areas receiving blood from these distant grafts with extensive distributions (16/24), and left ventricular wall motion was preserved or significantly improved postoperatively in 28/31 segments in such areas. When comparing angiographic and scintigraphic methods of evaluating myocardial perfusion, the angiogram underestimated the full extent of graft blood flow distribution in 13 of 24 instances (54 percent).

The effect of vasodilator therapy on systolic and diastolic time intervals in congestive heart failure.

To determine the effects of vasodilator drugs on systolic and diastolic time intervals in patients with congestive heart failure, these noninvasive measurements were made in 50 patients with moderately severe disease before and after vasodilator therapy. These data were compared with direct hemodynamic measurements obtained simultaneously from a flow-directed triple-lumen pulmonary artery catheter placed in each patient. Oral vasodilators included isosorbide dinitrate 20 mg (n-8), hydralazine 75 mg (n-8), and 100 mg (n-8), and first-dose prazosin 2 mg (n-8) and 5 mg (n-8). Ten additional patients received intravenous dose-response infusions of isosorbide dinitrate, nitroglycerin, and nitroprusside. Increases in cardiac output and stroke volume effected a decrease (improvement) in the systolic time interval parameter of ventricular performance, PEP/LVET. The correlation coefficient for vasodilator-induced changes in cardiac output vs changes in PEP/LVET was -0.82 (P less than 0.05) and for changes in stroke volume and PEP/LVET was -0.75 (P less than 0.05). In general, the diastolic time period was not significantly affected by these drugs. The systolic time intervals are a simple and useful method for evaluation of vasodilator therapy in patients with congestive heart failure.

Negative inotropic effect of lidocaine in patients with coronary arterial disease and normal subjects.

The effect of administration of lidocaine on left ventricular performance was studied using systolic time intervals in nine normal subjects, eight patients with stable angina, and 15 patients with acute myocardial infarction. The greatest response in systolic time intervals occurred at three minutes after intravenous injection of lidocaine (100 mg), with values returning to baseline at 10 to 15 minutes. Administration of lidocaine produced a significant prolongation of the preejection period (PEP) corrected for heart rate in all groups and a prolongation of the ratio of PEP to left ventricular ejection time (PEP/LVET) in patients with angina. The group with acute myocardial infarction exhibited a hyperadrenergic state, as shown by a short baseline QS2I. The QS I was lengthened by administration of lidocaine in all groups, but this was more profound in those with acute myocardial infarction. These changes in systolic time intervals were still present at two hours after injection in six patients with acute myocardial infarction in whom an infusion of lidocaine followed the initial bolus. The effect of administering lidocaine after intravenous injection of propranolol (5 mg) was also studied in six normal subjects. Although propranolol therapy along prolonged the PEP/LVET, a further significant prolongation followed subsequent injection of lidocaine.

Systolic time intervals in atrial fibrillation.

Forty patients with atrial fibrillation and 20 patients with congestive heart failure and sinus rhythm were studied. Patients were divided into two groups. Group A consisted of 20 patients with atrial fibrillation in whom systolic time intervals were measured. Twenty to 50 beats were analyzed. Five of the patients had high-fidelity measurements of left ventricular pressure simultaneous with determination of systolic time intervals. Analysis of the systolic time intervals for the entire group showed that the preejection period lengthened at faster heart rates and that the left ventricular ejection time was relatively constant at slower heart rates. This resulted in a progressive increase in the ratio of preejection period over left ventricular ejection time (PEP/LVET) as the heart rate increased. The rate of increase in PEP/LVET was minimal below a heart rate of 75 beats per minute. The increase in preejection period at faster heart rates is due to greater isovolumic developed pressure without a corresponding increase in left ventricular dp/dt. Group B consisted of 40 additional patients (20 with atrial fibrillation and 20 with sinus rhythm). In group B, the total electromechanical systole corrected for heart rate (QS2I) and the levels of digoxin in the blood were compared. The QS2I was significantly shorter in atrial fibrillation (497 +/- 5 msec vs 528 +/- 4 msec; P less than 0.01), while the levels of digoxin in the blood were identical (0.9 +/- 0.1 vs 1.0 +/- 0.1 ng/ml). The results of this study must be considered when systolic time intervals are to be employed in patients with atrial fibrillation.

Repolarization variant vs acute pericarditis. A prospective electrocardiographic and echocardiographic evaluation.

Patients with ECG nonischemic ST segment elevation were prospectively studied to determine the accuracy of the initial ECG diagnoses. Evaluations were made of 131 consecutive patients by serial clinical, ECG, and echocardiography to establish a diagnosis. Eighty-six (66 percent) had an initial ECG interpretation of repolarization variant. Only three of the 86 (3 percent) subsequently met clinical criteria for acute pericarditis. Analysis of the mean frontal ST segment vector and PR segment depression revealed no significant differences between the 119 patients with final clinical diagnosis of repolarization variant and the 12 with clinical acute pericarditis. The diagnostic reliability of the initial ECG alone as a means of confirming acute pericarditis is low (9/45, 20 percent) but in the detection of repolarization variant is extremely high (83/86, 97 percent).

Time couse of the blockade effect of propranolol on sinus node and atrioventricular node.

The time course of the blockade effect of propranolol on the sinus node (SN) and the atrioventricular node (AVN) was studied in six normal volunteers. Serial isoproterenol infusions were done before and after oral propranolol administration, 160 mg daily for two days. The inhibition by propranolol of the heart rate increases due to isoproterenol was used to assess the blockade of the sinus node, and the diminution by propranolol of the shortening in the PR interval due to isoproterenol was used to assess the blockade of the atrioventricular node. The blockade effects on the sinus node and the atrioventricular node were identical and persisted more than 24 hours. There was no good relationship between plasma propranolol and blockade effect on sinus node and/or atrioventricular node as propranolol was no longer detectable in the plasma 24 hours after the last dose. A transient hypersensitivity to isoproterenol was present 36 to 48 hours after propranolol withdrawal. The explanation of these phenomena most likely lies in the peculiar nature of beta-adrenergic receptors.

Extension of the serum digoxin concentration--response relationship to patient management.

The purposes of this investigation were to demonstrate how computer simulations may be employed to extrapolate data obtained from a single intravenous digoxin dose to multiple oral dosing patterns and how these simulations may apply to clinical situations. The intravenous data were obtained from a previous study of the pharmacokinetics of serum digoxin and its inotropic response (derived from systolic intervals) in 12 normal male volunteers. The simulations were applied to various clinical situations including variations in oral dosing, alternate loading doses, no loading versus loading dose, and intravenous versus oral dosing. A nonlinear relationship was found between response and the post-distribution serum digoxin concentration in the therapeutic range. Thus, the increase in inotropic response is less than proportional to the increase in digoxin concentration in serum. This nonlinear relationship has several important clinical implications for loading and maintenance dosing protocols. Such concepts may be important relative to more rational clinical use of digoxin and to decreasing digoxin toxicity.

Safety and rationale for continuation of propranolol therapy during coronary bypass operation.

Thirty consecutive patients undergoing coronary bypass were studied. Oral propranolol therapy was maintained up to 4 to 10 hours before operation. Nineteen of the patients had a history of myocardial infarction (MI), 14 had hypoakinetic areas, and 8 had decreased ejection fraction indicating advanced coronary artery disease. Twenty-four-hour urinary epinephrine and norepinephrine obtained the day before operation were markedly increased at 136 +/- 12 microgram per 24 hours (normal, 39 +/- 4 microgram, rho less than 0.01). There were 4 perioperative MIs (13%) and no deaths. Plasma propranolol 1 hour before operation was 43.3 +/- 8 ng per milliliter, indicating good beta blockade. Propranolol was started within 24 hours postoperatively. There were no preoperative, intraoperative, or postoperative complications related to propranolol therapy. We conclude that because of markedly increased adrenergic tone the day before operation and transient hypersensitivity to adrenergic stimulation after withdrawal of propranolol, this agent should be continued through coronary bypass operation.