THE CHANGING FACE OF AIDS

Patients with acquired immunodeficiency virus (AIDS) who are admitted to an intensive care unit (ICU) increasingly present with bacterial sepsis rather than with Pneumocystis carinii pneumonia (PCP), according to the authors of an observational cohort study conducted in Washington, DC. AIDS patients who have a sepsis syndrome of unknown etiology should receive broad-spectrum antibacterial therapy.

Rosenberg and colleagues conducted a 2.5-year study of the causes and outcomes of sepsis in 129 consecutive AIDS patients admitted to an ICU. In 102 patients (79%), the primary reason for admission was infection; 46 (45%) of the patients had bacterial infections, with Pseudomonas aeruginosa, Staphylococcus aureus, and Streptococcus pneumoniae the most commonly isolated organisms. The lung was the most common site of infection. Lung infections were responsible for 66 (65%) of the admissions; PCP, in 27 patients (41%), was the most common cause of pneumonia.

Overall hospital mortality was 54%. It was much higher—68%—among the patients with bacterial sepsis. Eighteen percent of the patients had sepsis and 73% had septic shock. Mortality correlated significantly with the severity of infection and of sepsis. The most important single clinical predictor of mortality was the admission APACHE III score. Bacterial infections and pneumonia were also predictive of mortality.

The authors suggest that patients who are not receiving or no longer responding to highly active antiretroviral therapies may be at increasing risk for bacterial infections. Thus, greater attention should be paid to bacterial sepsis when evaluating AIDS patients admitted to an ICU.

In an accompanying editorial, Proctor notes that in patients with AIDS, infections may progress from routine to more severe and thus may require ICU admission. In view of the high mortality associated with bacterial sepsis in the study by Rosenberg and colleagues, he recommends that broad-spectrum antibiotics be used in all seriously ill human immunodeficiency virus–infected patients until it becomes possible to exclude infections with pyogenic bacteria.

Proctor adds that combination antibiotic therapy will be needed to treat drug-resistant S aureus and P aeruginosa infections. A more aggressive approach at the onset of antibiotic therapy may reduce mortality.

Rosenberg AL, Seneff MG, Atiyeh L, et al. The importance of bacterial sepsis in intensive care unit patients with acquired immunodeficiency syndrome: implications for future care in the age of increasing antiretroviral resistance. Crit Care Med. 2001;29:548-556.

Proctor RA. Bacterial sepsis in patients with acquired immunodeficiency syndrome [editorial]. Crit Care Med. 2001;29:683-684.

10-DAY STEROID REGIMEN SUPERIOR TO THREE-DAY REGIMEN IN COPD

In patients with severe exacerbations of chronic obstructive pulmonary disease (COPD), a 10-day course of corticosteroids is superior to a three-day course in terms of outcome but not exacerbation rates, report the authors of a prospective, randomized study from Turkey.

Sayiner et al compared the effectiveness of three- and 10-day courses of methylprednisolone in 36 COPD patients who had experienced exacerbations requiring hospitalization. For the first three days, all patients were treated with methylprednisolone (0.5 mg/kg every six hours). Eighteen patients were then randomized to continue treatment for an additional seven days, during which the dose was tapered and eventually discontinued. The other 18 patients received placebo after the first three days. Seventeen patients in each group completed the study.

Arterial oxygen tension and forced expiratory volume in one second improved significantly in both groups, but more so in patients treated for 10 days. The pH levels normalized in both groups, and the arterial carbon dioxide tension did not change in either group. Forced vital capacity improved only in patients on the 10-day regimen. Both groups reported significant improvement in all symptom scores (dyspnea during the day, at night, and with exertion; cough; and sputum volume), but dyspnea with exertion was significantly more improved in the 10-day group. Hyperglycemia developed in two patients in each group and was controlled by adjusting caloric intake.

During a six-month follow-up, six patients in the three-day group experienced eight exacerbations, and the five patients in the 10-day group had one exacerbation each. These differences were not significant.

Sayiner A, Aytemur ZA, Cirit M, Ünsal I. Systemic glucocorticoids in severe exacerbations of COPD. Chest. 2001;119:726-730.

LOW-DOSE VASOPRESSIN VERSUS HIGH-DOSE CATECHOLAMINES

Low-dose infusions of vasopressin are associated with clinically important increases in blood pressure, systemic vascular resistance, and urinary output in hypotensive patients with vasodilatory septic shock, say the authors of a prospective, case-controlled trial from Japan.

Tsuneyoshi et al studied the cardiovascular and metabolic effects of a continuous intravenous infusion of vasopressin (0.04 U/min for 16 hours) in 16 septic patients with persistent hypotension despite adequate fluid resuscitation and infusions of pharmacologic doses of catecholamines. Vasopressin therapy resulted in significant increases in blood pressure and systemic vascular resistance index, starting with the first measurement at two hours and continuing to the end of treatment. No significant changes were seen in mean pulmonary artery pressure, pulmonary vascular resistance, cardiac index, heart rate, pulmonary artery occlusion pressure, or central venous pressure.

Vasopressin therapy stabilized 14 of the 16 patients. Two remained unstable with persistent hypotension. Overall, there were no significant treatment-related changes from baseline in the metabolic or electrolyte values or in the concentrations of atrial natriuretic peptide, aldosterone, angiotensin II, or renin. Urinary output increased in 10 patients, but not in six patients who were anuric before the start of treatment. The authors noted that this effect may be nonspecific.

Seven patients died, for a mortality rate of 44%. They included the two patients who remained hypotensive despite vasopressin therapy and five patients with multiple organ failure.

Tsuneyoshi I, Yamada H, Kakihana Y, et al. Hemodynamic and metabolic effects of low-dose vasopressin infusions in vasodilatory septic shock. Crit Care Med. 2001;29:487-493.

CONTINUOUS-INFUSION AMPHOTERICIN B REDUCES TOXICITY, RETAINS EFFICACY

Amphotericin B is the mainstay of treatment for immunocompromised patients with life-threatening fungal infections; however, treatment is associated with infusion- and dose-related adverse reactions. Administering amphotericin B by continuous infusion reduces its toxicity without decreasing its effectiveness, say the authors of a prospective, controlled trial from Switzerland.

Eriksson et al randomized 40 patients with refractory fever and suspected or proved invasive fungal infections to a continuous (24-hour) infusion of amphotericin B, and 40 similar patients to rapid (four-hour) infusion. Both groups received a median dose of about 0.96 mg/kg. Thirty-seven and 36 patients, respectively, were severely neutropenic before treatment. The patients were followed for three months after the end of treatment.

The continuous-infusion patients experienced fewer side effects and, therefore, were less likely to be treated for fever or chills after the first treatment day. C-reactive protein levels, which were similar in the two groups at baseline, rose significantly in the rapid-infusion group at 24 and 48 hours. Creatinine clearance was significantly less impaired in patients undergoing continuous infusion during and at the end of treatment.

Treatment had to be discontinued in two patients in the rapid-infusion group because one had refractory leukemia and the other severe nephrotoxicity, and in one patient in the continuous-infusion group because of refractory leukemia. Twelve patients in the rapid-infusion arm and four patients in the continuous-infusion group died by three-month follow-up. Seven deaths occurred during the treatment period; all were in the rapid-treatment group.

Eriksson U, Seifert B, Schaffner A. Comparison of effects of amphotericin B deoxycholate infused over 4 or 24 hours: randomised controlled trial. BMJ. 2001;322:579-582.

C PNEUMONIAE INFECTION MAY LIMIT AIRFLOW IN ASTHMATIC PATIENTS

Patients with nonatopic adult-onset asthma may be at risk for persistent airflow limitation due to recurrent or chronic infection with Chlamydia pneumoniae, say the authors of a multicenter, cross-sectional study from the Netherlands.

Brinke et al examined the relationship between loss of lung function and seropositivity to C pneumoniae in 101 patients, ages 18 to 75 years (mean, 46 years), with severe asthma. Fifty-one percent had early-onset asthma (before age 18 years) and 49% had late-onset asthma (age 18 or older). Exclusion criteria included current smoking or a smoking history of more than 10 pack-years. Enzyme-linked immunosorbent assay of C pneumoniae–specific serum immunoglobulin (Ig) G and IgA antibodies showed a seroprevalence of 73% and 60%, respectively, with no significant differences overall between those with early- versus adult-onset asthma.

The annual loss of lung function was determined for various patient subgroups categorized according to age at onset, atopic status, and IgG seropositivity for C pneumoniae. This loss was calculated by dividing the ratio of forced expiratory volume in one second and vital capacity (both expressed as a percentage of predicted) by the duration of asthma.

Patients with adult-onset nonatopic asthma and positive IgG antibodies to C pneumoniae had about a four times greater loss of lung function than did the other subgroups. However, C pneumoniae IgA seropositivity was not associated with a significant differences in lung function in any subgroup.

Brinke A, van Dissel JT, Sterk PJ, et al. Persistent airflow limitation in adult-onset nonatopic asthma is associated with serologic evidence of Chlamydia pneumoniae infection. J Allergy Clin Immunol. 2001;107:449-454.

RAPID DIAGNOSIS OF PE BY D-DIMER ASSAY, ALVEOLAR DEAD SPACE

Vascular imaging of patients with suspected pulmonary embolism (PE) is expensive and time-consuming and may expose patients to ionizing radiation. The combination of a negative whole blood agglutination D-dimer assay and a normal alveolar dead-space fraction identifies those patients with a very low probability of PE, report the authors of a prospective, multicenter study.

Kline et al investigated the predictive value of a normal whole blood agglutination D-dimer assay combined with a normal alveolar dead-space fraction (volume of alveolar dead space/tidal volume of 20% or less) in excluding PE in 380 hemodynamically stable emergency department patients 18 years or older. All measurements were made at bedside before the patients underwent pulmonary vascular imaging, either ventilation-perfusion scintillation lung scanning or contrast-enhanced helical computed tomography of the chest. The patients were followed by telephone for about six months.

Sixty-four patients (16.8%) were diagnosed with PE. Of these, 20 had an abnormal D-dimer assay result, three had an abnormal dead-space fraction, and 40 had abnormal results in both tests. One patient (one of seven taking warfarin) had normal results in both tests, for a sensitivity of 98.4%. Of the 316 patients without PE, 163 had normal results in both tests, for a specificity of 51.6%.

The authors emphasize that while the alveolar dead-space fraction enhances the diagnostic accuracy of the D-dimer assay, the fraction was normal in almost one third of the patients with PE. Thus, the dead-space measurement should not be used as the only screening test for PE.

Kline JA, Israel EG, Michelson EA, et al. Diagnostic accuracy of a bedside D-dimer assay and alveolar dead-space measurement for rapid exclusion of pulmonary embolism: a multicenter study. JAMA. 2001;285:761-768.

AN OBJECTIVE METHOD FOR SCORING BURN-RELATED ORGAN DYSFUNCTION

Readily available objective clinical and laboratory data can be used to gauge the extent of burn-related organ dysfunction, according to the authors of a prospective study from Texas. These data may also be used as evidence of treatment effectiveness.

Cumming et al assessed organ dysfunction, organ failure, and sepsis in 85 patients with total body surface area burns of at least 20% hospitalized during a single year. Organ dysfunction and the severity of sepsis were determined by a modified version of the multiple organ dysfunction (MOD) score, which excludes the central nervous system component, but retains the pulmonary, renal, cardiovascular, hepatic, and hematologic systems.

Of the 85 patients, only 32 (37.6%) had no evidence of organ dysfunction. Sepsis occurred in 43 patients (50.6%); 12 patients (14.1%) had severe sepsis or septic shock. Only 32 patients had no evidence of organ dysfunction. Fifteen patients (17.6%) had dysfunction in two or more organ systems; 24 (28%) had severe MOD (defined as a MOD score of 6 or greater).

The risk for MOD correlated with age and total body surface area burned. MOD occurred more commonly in men than in women. In patients with and without severe MOD, mortality was 29.2% and 9.8%, respectively. Similarly, age and larger burn area were the most important factors associated with the development of severe sepsis/septic shock. Both severe MOD and severe sepsis/septic shock were related to male gender, as well as to prolonged intensive care unit stays and prolonged mechanical ventilation.

Cumming J, Purdue GF, Hunt JL, O’Keefe GE. Objective estimates of the incidence and consequences of multiple organ dysfunction and sepsis after burn trauma. J Trauma. 2001;50:510-515.

CLINICAL COURSE AFFECTS PNEUMONIA DEVELOPMENT

A prospective cohort study has determined that nosocomial pneumonia infection may not always stem from preadmission issues. It may be caused by specific factors associated with intensive care unit (ICU) treatment.

Tejada Artigas et al studied 103 critically ill patients who were consecutively admitted to a trauma ICU in Zaragosa, Spain. Upon admission, data were compiled on the patients’ injuries and preexisting conditions using the Glasgow Coma Scale, the Simplified Acute Physiology Score system, and the APACHE II scale.

Variables in the patients’ clinical courses were later observed on a daily basis, prior to the onset of nosocomial pneumonia in some instances. The mortality rate was 43.5% in the 23 patients (22.3%) who had contracted pneumonia while in the ICU compared with 18.8% in patients who had not. The authors said that pneumonia significantly increased the death rate among critically ill patients who might otherwise have seen more positive outcomes.

The main risk factor associated with pneumonia in the study patients was the use of prolonged mechanical ventilation with positive end-expiratory pressure. The researchers noted that use of endotracheal tubes in mechanical ventilation circumvents natural defenses against infection by generally allowing a more direct path to the respiratory tract for germs.

Corticotherapy, craniotomy, and continuous enteral feeding were also strongly related to development of pneumococcal infection. Intensifying hygienic and prophylactic measures in ICU patients at increased risk for nosocomial infection, the authors said, would aid in reducing pneumonia incidence and its costs and complications in ICUs.

Tejada Artigas A, Bello Dronda S, Chacón Vallés E, et al. Risk factors for nosocomial pneumonia in critically ill trauma patients. Crit Care Med. 2001;29:304-309.