Prediction of staphylococcal etiology among patients with septicemia with or without endocarditis by multivariate statistical methods

Multivariate statistical methods, multiple regression (RA) and automatic interaction detector analysis (AID) were used to study the possibility of an early prediction of staphylococcal etiology in 249 of 851 patients with verified septicemia or endocarditis. The variables included pertinent symptoms and signs and laboratory data available soon after admission. 10 of the 70 variables initially studied showed simple, or in various combinations, a statistically significant partial correlation to staphylococcal etiology in the AID. The highest predictive value with a high probability for staphylococcal etiology was recorded for combinations of the variables: i.v. narcotic addiction and septic pulmonary embolism; non-addiction, wound infection, and hospitalization within 4 weeks; non-addiction, absence of skin infection, presence of foreign body, and age less than 60 yr. Staphylococcal etiology was contradicted by the absence of i.v. narcotic addiction, skin infection, foreign body, septic skin manifestation, surgical procedure within 4 weeks, joint symptom and a C-reactive protein less than or equal to 10 mm. Thus, a prediction of etiology may be valuable in choosing therapy before definite confirmation by positive blood cultures or when blood cultures remain sterile.

Intravenous drug addiction--staphylococcal septicemia--pulmonary embolism: a triad pathognomonic for tricuspid valve endocarditis?

28 patients (29 episodes) presenting the triad of intravenous narcotic drug addiction, Staphylococcus aureus septicemia and septic pulmonary embolism were examined for the presence of tricuspid valve endocarditis. In a prospective study, 12/13 patients examined by echocardiography had vegetations on the tricuspid valve, in 6 detected by M-mode, in 5 by 2-D technique, and in one by both methods. In 4 patients with tricuspid endocarditis, no significant murmurs were heard. Similarly, in a retrospective study, 5/16 patients not subjected to echocardiography had no murmurs consistent with tricuspid valve endocarditis. One of these patients died. In 2/16 patients the diagnosis was verified at autopsy and in 1 at heart surgery. In both groups, altogether 11 patients never developed murmurs, in 7 of them despite demonstrable vegetations. Therefore, we consider the triad to be pathognomonic for tricuspid valve endocarditis, provided that no peripheral source of emboli is present. Consequently, patients fulfilling these criteria should be treated as suffering from endocarditis. Echocardiographic examination does not seem necessary in patients with a typical clinical picture but may be helpful when chest X-rays are inconclusive or complications are suspected.

A prospective study on septicemia. II. Clinical manifestations and complications, results of antimicrobial treatment and report of a follow-up study

In a prospective study at a hospital for infectious diseases 151 patients, 110 with proved and 41 with probable septicemia, were analysed. Clinical, laboratory, therapeutic data and prognostic experiences, partly from a follow-up study, are described. Secondary manifestations, mostly from skin, mucous membranes, nervous system and lungs, were present in 72%. They were more often caused by gram-positive cocci than by gram-negative baccilli and in some cases not revealed until autopsy. Lesions in the nervous system were most often caused by strepto- or penumococci or Haemophilus influenzae. In 2 splenectomized patients with extensive hemorrhages, pneumococci were isolated. Subacute courses were rare even in alpha-streptococcal infection and its "classical signs" were never observed. Shock and thrombocytopenia suggesting disseminated intravascular coagulation occurred together in 11%, and in one-third in the lethal cases. Gram-positive bacteria were often involved. Leukocytosis was absent in 53 patients; 20 were alcohol or narcotic drug abusers, and 7 died. ECG changes were registered in 33%. Initial antibiotic treatment was applied according to a fixed schedule, with cure in 61% on this first treatment, and especially so in infections with gram-positive cocci. During the initial hospital stay 20% died from uncontrolled infection. All had underlying diseases or factors, often major causes of death. The infection was regarded as hospital-acquired in 40% among the lethal cases. During a one-year follow-up period 3 patients died from a new septicemia and 10 from their underlying disease.

Haemophilus parainfluenza--an uncommon cause of septicemia and endocarditis

6 cases of endocarditis and 1 of septicemia caused by Haemophilus parainfluenzae have been observed in our hospital from 1970 to 1977, as against no case from 1957 to 1969. The mean age of the patients was 46 years. The clinical picture did not differ from that seen in cases of septicemia and endocarditis from other cases. In 4 cases no underlying heart disease was known. In 2 of them, endocarditis developed in the mitral and in 1 in the aortic valve. Of 3 patients with preexisting heart disease, 2 had involvement of the aortic valve and 1 of the mitral valve. Six patients were cured, 2 or possible 3 by treatment with ampicillin, 2 with cephalothin, and 1 with co-trimoxazole. In 2 patients intractable heart failure necessitated the insertion of prosthetic valves, and 1 patient died. Thus, cases of septicemia and endocarditis due to H. parainfluenzae have been observed only in recent years and they appear to be serious infections.

Septicemia I. A prospective study on etiology, underlying factors and sources of infection

In a prospective study, 110 cases of proved and 41 of probable septicemia were analysed. Underlying diseases or factors were common, also in young patients, often narcotic drug or alcohol abusers. In one third, septicemia was probably hospital-acquired. Streptococci, staphylococci and gram-negative enteric rods were most common, followed by pneumococci and haemophilus species. Staphylococci and gram-negative enteric rods dominated in patients hospitalized at or before the onset of disease, and staphylococci in patients with vascular foreign bodies and in drug addicts. Enteric rods were common in elderly patients with urogenital disease, especially after instrumentation. Among streptococci, alpha-streptococci dominated; they did not emanate as often from dental foci as expected. In 2 of 3 asplenic patients with extensive hemorrhages pneumococci were found. Bacteria of low virulence and fungi occurring as opportunists were rare. Mixed infections, present in 4 cases, were severe. Portals of entry could often be proved or assumed. In half of all patients, the same organism was isolated from a primary focus as from blood. The importance of underlying illnesses and factors and of invasive procedures was evident. A tentative etiological diagnosis could often be made, based on case history, underlying factors and the probable portal of entry.

Cefamandole nafate: an evaluation of antibacterial activity, serum levels, clinical effect, and incidence of side reactions in 58 patients

35 males and 23 females with skin and soft tissue infection, lower respiratory tract infection, urinary tract infection or septicemia with known etiology were treated with cefamandole nafate. The patients were to a large extent (30%) alcoholics and/or drug abusers. Cefamandole was given intravenously as 4 daily doses of 1 g in 52 cases and of 2 g in 6 cases for 8 to 16 (mean 10) days. The effect was considered to be good in 40 patients (70%). Adverse reactions, mostly slight and transient, were seen in 22 patients (38%). Peak serum levels varied from 26 to 82 (mean 50) micrograms/ml after 1 g doses and from 68 to 100 micrograms/ml after 2 g. Previous statements of a better in vitro activity of cefamandole than of older cephalosporins against some gram-negative bacilli were corroborated.

Bacterial endocarditis. II. A prospective study with clinical, laboratory and therapeutic observations

In a prospective study 31 cases of proved and 10 of highly suspected bacterial endocarditis were analysed. Valve incompetence was the usual consequence but stenosis occurred in 3 cases, all fatal. Congestive heart failure developed in two-thirds. Secondary manifestations were common and as often caused by alpha-streptococci as by other bacteria. Initial antibiotic treatment was mainly applied according to a fixed schedule, generally with continuous intravenous infusion, followed by oral therapy. In 10 patients, the infection was still active after 6 weeks of therapy. Therefore, we now use intermittent injections or infusions for at least 4--6 weeks. Within a year, 9 patients died from uncontrolled infection together with congestive heart failure, and 1 from heart failure and active chronic endocarditis. In 4, myocardial abscesses or inflammations were found. All 10 had underlying factors or advanced stages of the disease. Of 18 patients with alpha-streptococci or enterococci none died from endocarditis, as against 10 of 23 with other or unknown bacteria. Follow-up yielded valuable information on one-third of the patients. The mortality during the initial hospital stay was 22%, after 1 year 24% and after 5 years 39%.

Bacterial endocarditis. I. A prospective study of etiology, underlying factors and foci of infection

In a prospective clinical and bacteriological study of 192 consecutive patients with septicemia admitted during the period 1967--1971 to a hospital for infectious diseases, 31 cases of proved (group I) and 10 with probably (group II) bacterial endocarditis were observed and analysed with regard to a variety of factors. The incidence of bacterial endocarditis was 3 per 1 000 admissions. The mean age of the patients was 52.6 years; there was no sex predominance. In about one third of the patients apparently normal valves were involved. Underlying non-cardiac factors were present in two thirds, even in younger age groups, in which chronic alcoholism and intravenous drug abuse were frequent. Secondary manifestations from various organs were noted in about three quarters of the patients, in one quarter already on admission. alpha-Hemolytic streptococci and staphylococci were most commonly isolated, followed by gram-negative enteric rods, beta-hemolytic streptococci, enterococci, and pneumococci. A probable portal of entry could be assumed in 80% of the patients and could often be related to the type of bacteria involved. In some cases, diagnostic or therapeutic procedures preceded the disease. A subacute course of the endocarditis was observed in one third of the patients and all these had alpha-hemolytic streptococci or enterococci. The aortic and mitral valves were equally often involved and equally often infected by alpha-hemolytic streptococci. In 4 patients, 3 of whom were drug addicts, the tricuspid valve was probably involved; all were infected by staphylococci. In one case autopsy revealed a mural endocarditis.

Two-stage replacement of tricuspid valve in active endocarditis

One month after a 21-year-old female drug addict with tricuspid staphylococcal endocarditis seemed to have been cured by antibiotic therapy she relapsed. Further antibiotic treatment for one month failed to control the infection. Tricuspidectomy was when performed and the infection was eliminated. A valve prosthesis was inserted 3 months later. She recovered and is now fully restored to health. We recommend this two-stage surgical technique in similar cases.