Infectious Diseases: February 15, 2004

John Bartlett, MD


February 17, 2004

In This Article

Advances in Diagnostic and Surveillance Methods

Dickson SJ, Brent A, Davidson RN, Wall R. Comparison of bronchoscopy and gastric washings in the investigation of smear-negative pulmonary tuberculosis. Clin Infect Dis. 2003;37:1649-53. This is a report from the Northwick Park Hospital in the United Kingdom comparing the yield of Mycobacterium tuberculosis in cultures of gastric washings vs bronchoscopy in patients with smear-negative pulmonary tuberculosis (TB). This was a retrospective analysis of patients with the following criteria: (1) must have had both bronchoscopy and gastric washings, (2) negative results of acid-fast bacilli (AFB) smears of sputa; and (3) subsequent treatment for TB. There were 180 patients who satisfied these criteria, including 141 immigrants who were referred from Heathrow and Gatwick Airports due to abnormal test x-rays. The procedure also included a post-bronchoscopy expectorated specimen. The results showed that the yield with bronchoalveolar lavage (BAL) was significantly better than with gastric washings. As expected, the more procedures done, the higher the total yield. The results are summarized in Table 9 .

The authors conclude that bronchoscopy is superior to gastric washings in the diagnosis of smear-negative pulmonary TB. They recommend the combination of bronchoscopy and 2 gastric washings as the optimal method for diagnosis of smear-negative TB.

Comment: The post-bronchoscopy specimen became popular many years ago when there was concern that the lidocaine used in bronchoscopy would interfere with the growth of M tuberculosis since it was active against the organism in vitro. For those who still get post bronchial specimens, it is noteworthy that the yield of the BAL was approximately 2-fold higher and only slightly increased when both were done (34% vs 30%). For gastric washings, the yield depended on the number done, but a single BAL was significantly better than 3 washings (30% vs 21%). The authors advocate a combination of BAL plus post-bronchoscopy sputa and 2 gastric washings as the optimal method, although the 2 gastric washings added only 3% to the total yield (34% vs 38%). If bronchoscopy is not possible, they recommend 3 gastric washings, which gave a yield of 21%.

Raad I, Hanna HA, Alakech B, et al. Differential time to positivity: a useful method for diagnosing catheter-related bloodstream infections. Ann Intern Med. 2004;140:18-25. The study was done at MD Anderson Cancer Center in Houston to determine whether the difference in time to positive blood culture results between blood from peripheral vein vs central catheter could be used to accurately distinguish catheter-related bloodstream infections (CRBIS). Quantitative cultures were done as well. Care was taken to inoculate the media simultaneously. Catheters were removed aseptically if infection was suspected or if they were not needed. The 5-cm segment was cultured by semiquantitative roll-plate method. The study included 6138 simultaneous blood cultures, and the analysis was restricted to 191 who had positive results for the same organism in both the peripheral and catheter cultures. The breakdown of the results of cultures is shown in Table 10 .

The definition of catheter-related bacteremia was adopted from the Infectious Diseases Society of America guidelines[22] based on a significant catheter-tip colonization (at least 15 colony-forming units [CFUs]) for the same organism in both cultures or quantitative blood cultures that yielded at least 5-fold greater bacteria in blood drawn through the central venous catheter. The results are shown in Table 11 based on a time differential for positive growth that was greater or less than 120 minutes.

The authors conclude that cultures become positive more rapidly from catheter-drawn specimens with line-associated sepsis, and 120 minutes is the threshold for this difference that shows a high degree of sensitivity and specificity.

Comment: There are several important observations in this study, including some that were pointed out in the editorial by Barry Farr from Virginia[23]:

  1. The issue has important and practical application in patients with long-term catheters due to the difficulty and expense for replacement. Thus, the ability to avoid unnecessary catheter removal is highly desired.

  2. It is important to note that among the 6000 cases with suspected catheter-related bacteremia, only 3% had this diagnosis verified.

  3. The differential time to positive results applies only to patients with positive results in both cultures, with the assumption that media were inoculated simultaneously with specimens from both sources.

  4. Blood cultures that yield Candida or Staphylococcus aureus are likely to be true positives, but one or even both cultures that grow coagulase-negative staph are often contaminates. This applies when 2 cultures grow this organism based on molecular typing.[24,25]

  5. The results in this study showed that specificity of time differential was lower in a subset of patients who had received antibiotics. This result is unexplained.

  6. Multiple studies have shown that simultaneous quantitative blood cultures yield 5-10 times the number of colonies in specimens from the catheter vs peripheral vein in catheter-related infections.[26] The method described here represents an alternative to this tedious and expensive quantitative method that is infrequently offered by clinical laboratories.

K pneumoniae

Paterson DL, Ko WC, Von Gottberg A, et al. International prospective study of Klebsiella pneumoniae bacteremia: implications of extended-spectrum beta-lactamase production in nosocomial infections. Ann Intern Med. 2004;140:26-32. The authors examine the frequency and risk factors for Klebsiella pneumoniae producing extended-spectrum beta-lactamases (ESBLs) using an analysis of 455 cases of bacteremia in 12 hospitals in 7 countries. Production of ESBL was determined with standard in vitro techniques from the National Committee for Clinical Laboratory Standards, and pulsed-field gel electrophoresis was used to determine genotypic relationships of the isolates. The results showed the following overall frequencies:

  • ESBL-producing K pneumoniae: 85/455 (18.7%)

  • nosocomial bacteremia: 78/253 (30.8%)

  • isolates from intensive care units: 30/69 (43.5%)

Frequency by country: South Africa - 31%, Taiwan - 19%, Australia - 17%, Argentina - 14%, United States - 10%, Belgium - 5% and Turkey - 4%

Mortality: 61/253 (24%) with no significant difference between those with and without ESBL-producing strains (27% vs 23% for nosocomial cases)

Risk for ESBL: The predominant identifiable risk was exposure to antibiotics containing the oxyimino group (cefuroxime, cefotaxime, ceftriaxone, ceftazidime, or aztreonam): The odds ratio for this exposure was 3.8.

Transmission: Patient-to-patient transmission correlated by hospital infection control policies. Six of 7 hospitals that did not routinely use contact isolation for these organisms had at least 2 isolates with the same genotype, but 3 of 5 hospitals that routinely use these precautions did not detect any patient-to-patient transmissions.

The authors conclude that K pneumoniae with ESBL is a "widespread nosocomial problem that requires appropriate infection and antibiotic management strategies."

Comment: The authors compare these organisms to vancomycin-resistant enterococci (VRE) and MRSA, but point out that physician awareness is substantially less because many hospitals do not routinely report the production of ESBLs. Their recommendation is that all K pneumoniae isolates should be routinely tested and that those that are positive should not be reported as susceptible to third-generation cephalosporins or cefepime. Another important factor in control is a reduction in patient-patient transmission by routine contact isolation.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.