To the Editor:
Septicemia is a life-threatening condition accompanied by a high mortality rate, especially in critically ill patients.1 Blood cultures are recommended to identify the microorganism responsible for infection before initiation of antibiotic treatment.2 Currently, automated blood culture systems are the standard detection methods. However, these are associated with a high rate of false-negative results due to inappropriate growth conditions or growth inhibition due to antibiotic treatment, a long time-to-result for detection and identification, requiring at least 2 days or, for fastidious organisms, even longer. A resulting delayed or inadequate treatment is associated with negative outcome.3
To overcome these limitations, newer technologies, i.e., polymerase chain reaction (PCR)-based techniques have been developed and are currently undergoing validation studies.4,5 While using such a technique, one assumes a quantitative relationship between bacterial DNA load, reflecting bacterial cell load in blood samples, and the number of PCR cycles needed for detection. Herein, we present an example of a patient in whom serial blood cultures were collected during initiation of antibiotic therapy along with results from in parallel eubacterial real-time PCR. Adequate antibiotic treatment, as confirmed later by microbiological testing, was observed to be associated with decreasing bacterial DNA load in the blood samples over time.
A 50-yr-old male patient (75 kg, 177 cm) with multiple myeloma developed high fever 6 days after human leukocyte antigen-identical stem cell transplantation from his sister. At this time, the patient received levofloxacin (500 mg IV per day). In addition, posaconazole (600 mg postoperatively, per day) and acyclovir (5 mg/kg IV) were administered. The first blood culture (each 5 mL) was taken in the evening by peripheral venous puncture, and meropenem (1 g every 8 h) was started immediately (Timepoint a). The next blood culture was obtained from a central vein on the next morning before the second infusion of meropenem (Timepoint b). About 1 h later, a third blood culture was taken (Timepoint c). By conventional blood culture, Staphylococcus epidermidis (oxacillin sensitive and levofloxacin resistant) was isolated in the first (a) and the second sample (b: both aerobic and anaerobic bottles), whereas the third pair (c) remained negative. In more detail, blood culture a became positive after 11.87 (aerobic) and 10.87 h (anaerobic), blood culture pair b after 15.54 and 13.87 h, respectively. The longer time to positivity of culture b compared with a and the negative result of culture c indicated growth inhibition of the infectious agent as an effect of antibiotic treatment.
For the detection of pathogens by PCR, 1-mL duplicates of whole blood were used with the SepsiTest™ kit (Molzym, Bremen, Germany). The kit involves lysis of human cells and degradation of human DNA by a DNase. DNase as well as a broad-spectrum 16S rDNA real-time PCR mix, which was run using an Opticon PCR cycler (Bio-Rad, Hercules, CA).5 Sequence analysis (www.sepsitest-blast.net) of the amplicons assigned the detected infectious agent to S. epidermidis (99% sequence identity).
Figure 1 shows the amplification curves of the PCR amplification of S. epidermidis DNA extracted from blood samples. Interestingly, positive results were obtained with all samples a through c. There was a shift to indicating later cycles of positivity, i.e., a lesser concentration of DNA in the sample, from a over b to c, all occurring before the negative control (d). The melting curve analysis confirmed the presence of bacterial DNA by showing clear peaks at melting temperatures 87.0°C-87.2°C (Fig. 2, a–c), whereas the negative control lacked a specific signal (d). These results indicated decreasing bacterial DNA loads in the blood after sampling and thus the time of antibiotic treatment. PCR results agreed with the blood culture results, indicating successful antibiotic treatment (see above).
Although PCR has supported the data from conventional blood cultures, it may yield results significantly earlier. Real-time PCR results can be obtained within approximately 4 h (DNA extraction and PCR), providing information at that time about a bacteremia or fungemia. Identification of the infectious agent requires approximately 3 additional hours (assuming a modern sequencing instrument is available in the laboratory).
A limitation of SepsiTest is that the presence of cell-bound bacterial DNA per se does not automatically mean viability of the pathogen. However, in our case, bacteria must have been viable, i.e., the increase in C(t) values observed with samples taken subsequently indicate a regression of cell numbers and thus loss of viability of S. epidermidis as an effect of antibiotic treatment.
In conclusion, this case demonstrates that the SepsiTest PCR kit may facilitate an accurate diagnosis of bacterial bloodstream infection. Moreover, the effect of adequate antibiotic treatment can be monitored as decreasing relative bacterial DNA load over successive blood sampling. Practical aspects for use of SepsiTest in a laboratory routine must await additional clinical data.
Samir G. Sakka, MD, PhD
Anna-Julia Kochem, MD
Department of Anesthesiology and Intensive Care Medicine
University of Witten/Herdecke
Medical Center Cologne-Merheim
Claudia Disqué, MD
High School Bremerhaven
Nele Wellinghausen, MD, PhD
Institute for Medical Microbiology and Hygiene
University Hospital of Ulm, Ulm, Germany
and Dr. Gaertner & Partner Laboratories
1. Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, Reinhart K, Angus DC, Brun-Buisson C, Beale R, Calandra T, Dhainaut JF, Gerlach H, Harvey M, Marini JJ, Marshall J, Ranieri M, Ramsay G, Sevransky J, Thompson BT, Townsend S, Vender JS, Zimmerman JL, Vincent JL. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Intensive Care Med 2008;34:17–60
2. Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel RP, Edmond MB. Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis 2004;39:309–17
3. Morrell M, Fraser VJ, Kollef MH. Delaying the empiric treatment of Candida
bloodstream infection until positive blood culture results are obtained: a potential risk factor for hospital mortality. Antimicrob Agents Chemother 2005;49:3640–5
4. Mühl H, Kochem A-J, Disqué C, Sakka SG. Activity and DNA contamination of commercial polymerase chain reaction reagents for the universal 16S rDNA real-time polymerase chain reaction detection of bacterial pathogens in blood. Diagn Microbiol Infect Dis (in press)
5. Wellinghausen N, Siegel D, Gebert S, Winter J. Rapid detection of Staphylococcus aureus
bacteremia and methicillin resistance by real-time PCR in whole blood samples. Eur J Clin Microbiol Infect Dis (in press)