Skip Navigation



Annals of Oncology Advance Access published online on September 5, 2007
Annals of Oncology, doi:10.1093/annonc/mdm351
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/11/1870    most recent
mdm351v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Penack, O
Right arrow Articles by Blau, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Penack, O
Right arrow Articles by Blau, I.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2007 European Society for Medical Oncology

Bloodstream infections in neutropenic patients: early detection of pathogens and directed antimicrobial therapy due to surveillance blood cultures

O Penack1,*, P Rempf1, M Eisenblätter2, A Stroux3, J Wagner2, E Thiel1 and IW Blau1

1 Department of Haematology, Oncology, and Transfusion Medicine
2 Institute for Microbiology and Hygiene
3 Department of Clinical Epidemiology and Statistics, Campus Benjamin Franklin, Charité, University of Medicine, Berlin, Germany

* Correspondence to: O. Penack, MD, Department of Immunology and Medicine Memorial Sloan-Kettering Cancer Center Zuckerman Building, 14th Floor, Room Z-1419, 1275 York Avenue, New York, NY, 10021, USA. Tel: +1-646-888-2317; Fax: +1-646-422-0452; E-mail: penacko{at}mskcc.org


   Abstract
Top
 Abstract
introduction
 patients and methods
 results
 discussion
 References
 
Bloodstream infections (BSIs) are frequent infectious complications in neutropenic patients. In order to determine the efficacy of surveillance blood cultures (BCs) to detect BSIs prior to clinical manifestation we performed a prospective trial. One hundred patients with haematological malignancies and long-term neutropenia following intensive cytotoxic therapies were recruited. BCs were taken thrice weekly during neutropenia. Forty-two patients were diagnosed with BSI. In 18 (43%) of those patients surveillance BC results were positive and identified microorganisms prior to onset of fever. In patients with positive surveillance BCs modification of the clinical management (specific antimicrobial therapy, CVC removal) resulted in a shorter time to defervescence (median 1.5 days) compared with patients with BCs positive after onset of fever (median 3.5 days, P = 0.004). In conclusion we detected causative microorganisms in more than one-third of BSIs prior to onset of clinical manifestation. The impact of surveillance BCs on the outcome has to be assessed in randomized studies.

blood culture, bloodstream, infection, neutropenia, surveillance


   introduction
Top
 Abstract
 introduction
patients and methods
 results
 discussion
 References
 
Bloodstream infections (BSIs) are one of the most serious complications in patients with haematological malignancies and profound neutropenia following intensive cytotoxic therapy. In this patient population BSIs are the most frequently documented microbiological infections [1–3]. BSIs are associated with severe complications such as septic shock and multi-organ failure, which are potentially life threatening, especially in immunocompromised hosts. The standard procedure for diagnosis of BSIs is to take blood cultures (BCs) at onset of fever, which is the clinical hallmark of BSIs in immune-competent hosts [4–6], but may be absent after intensive chemotherapy or haematopoietic stem cell transplantation (HSCT) due to the anergic state of the immune system. Several clinical trials showed improved survival for immediate empirical broad-spectrum antimicrobial therapy at onset of fever, which has become the standard procedure [7–9]. Despite such an immediate approach, BSIs in neutropenic patients often cause considerable morbidity and mortality. The high mortality is partly related to BSIs with microorganisms resistant to broad-spectrum antibiotics, which are emerging in many centres [10–13]. Thus, earlier detection and susceptibility testing of microorganisms causing BSIs may lead to optimization of antimicrobial therapy. In pilot studies we and others [14, 15] found that surveillance with BCs is feasible in neutropenic patients and helps in detecting significant microorganisms causing BSIs prior to the occurrence of signs of infections. Therefore we decided to perform a prospective trial with a larger number of patients and standardized diagnosis and treatment of infectious complications.


   patients and methods
Top
 Abstract
 introduction
 patients and methods
results
 discussion
 References
 
participants
According to the study protocol 100 patients were enrolled. All patients were recruited and treated at the department of Haematology, Oncology and Transfusion Medicine in the Charité-Campus Benjamin Franklin. Adult patients (≥18 years) with haematological malignancies, receiving chemotherapy likely to induce long-term neutropenia (neutrophil count <500/mm3 for >10 days) or undergoing HSCT were eligible. The study was approved by the Ethical Committee of the Charité and written informed consent was obtained from each patient before study entry. Subjects were ineligible for inclusion if they had clinical or microbiological evidence of systemic infection [e.g. bloodstream infection, pneumonia, fever of unknown origin (FUO)].

objectives
The objective of this trial was to assess the efficacy of surveillance BCs thrice weekly to detect infectious pathogens causing BSIs earlier than the standard procedure of BC sampling at onset of fever.

intervention
Blood sampling for surveillance BCs was performed every Monday, Wednesday and Friday during neutropenia. Blood was drawn from the central venous line (CVC) in all patients. BCs were drawn with a sterile system after a sterile pad was placed below the catheter hub and the hub was disinfected with 10% povidone–iodine. In cases of detection of pathogens in surveillance BCs, according to the study protocol, specific antibiotic therapy (in addition to broad-spectrum antibiotics, if applicable) was administered immediately at onset of clinical symptoms of BSI. In the absence of clinical symptoms of BSI no antibiotics were given in response to positive surveillance BC results.

outcomes
The primary endpoint was detection of the causative infectious pathogen in BCs drawn prior to onset of clinical symptoms of BSI. Criteria used to define BSI were those of the Centres for Disease Control (CDC) [16, 17]. In line with these definitions, BSIs were defined as isolation of a recognized pathogen from one ore more BCs that were unrelated to an infection at another site with or without fever or hypotension plus one of the following: isolation of a common skin contaminant in two separate BCs, isolation of a common skin contaminant in one BC in a patient with CVC and start of adequate antimicrobial therapy. When a pathogen was detected in surveillance BCs and the patient was diagnosed with BSI later on, the microorganism was considered to be the likely cause of the BSI under the following conditions: (i) absence of other foci of infection (e.g. pneumonia, urinary tract infection) plus detection of the same strain (with the same susceptibility pattern) in BCs at onset, or after onset, of clinical symptoms; (ii) absence of other foci of infection plus detection of the same strain (with the same susceptibility pattern) at the tip of the CVC at onset, or after onset, of clinical symptoms.

The number of false-positive surveillance BC results and the antimicrobial spectrum in BCs were analysed as secondary outcome variables. BC results were considered to be possibly false positive when: (i) the patient did not present with symptoms of BSI (including clinical deterioration without fever) within 5 days after a positive surveillance BC result, or (ii) the patient presented with clinical symptoms of BSI within 5 days after a positive surveillance BC result, but the microorganism detected in the surveillance BC was not identical with infectious pathogens identified at or after onset of clinical symptoms of BSI.

supportive care and diagnostic procedures
According to the standard clinical protocol in our institution all patients received oral topical amphotericin B four times daily plus antibacterial prophylaxis with levofloxacin 500 mg/day orally. Patients were examined daily for clinical signs of infection. Two separate sets of BCs (CVC and peripheral vein) plus a urine culture were obtained in all patients who developed fever. A chest X-ray was performed 1–3 days prior to neutropenia and at onset of fever. In all patients with atypical infiltrates in the chest X-ray, high resolution CT scan and brochoscopy were performed. Patients without lung infiltrates but persisting fever despite broad-spectrum antibiotic therapy underwent high-resolution CT scan within 5 days of the onset of fever. Patients with neutropenia and FUO immediately received broad-spectrum antibiotics. All patients without known intolerance received piperacillin–tazobactam (Tazobac®) 4.5 g three times daily. Treatment success was re-evaluated after 3 days. In cases of defervescence the regimen was continued for 7 days in total. In cases of persistent fever antibacterial therapy was changed to imipinem–cilastin (Zienam®) 1 g three times daily. A glycopeptide drug was added in patients with severe mucositis or suspected CVC infection. In all patients with suspected CVC infection the CVC was removed.

processing of blood cultures
Aerobic and anaerobic blood culture flasks were incubated at 37°C in a semi-automated blood culture system (BacT/Alert, bioMérieux, Marcy l'Etoile, France) for up to 7 days. If growth was indicated, subcultures for bacteria and fungi on solid culture media incubated in an aerobic CO2-enriched atmosphere as well as under anaerobic conditions were set up and preliminary identification by Gram staining was made. Identification using standard microbiological methods and semi-automated identification for Gram-negative bacilli (Vitek, bioMérieux) and susceptibility testing was performed according to DIN (Deutsches Institut für Normung) standards.

statistical analysis
Statistical analysis was performed using commercially available software (SPSS 14). Descriptives include absolute and relative frequencies for categorical data, median and range for numerical measurements (Tables 1 and 2). Due to the skewed distribution of time to defervescence, confirmatory comparison between patients presenting with BCs positive before and patients with BCs positive after onset of fever, and patients with negative BCs was performed using the Mann–Whitney U-test. In patients with multiple positive BCs, only the first positive surveillance BC and first positive diagnostic BC was used for statistical analysis. Diagnostic accuracy of surveillance BC outcome with regard to BSI and fever has been quantified in terms of positive and negative predictive values (PPV, NPV), sensitivity and specificity.


View this table:
[in this window]
[in a new window]

  		Table 1. Patient characteristics

 

View this table:
[in this window]
[in a new window]

  		Table 2. Isolated organisms in 71 positive BCs in 45 patients

 

   results
Top
 Abstract
 introduction
 patients and methods
 results
discussion
 References
 
patient characteristics
According to the study protocol 100 patients were recruited between January 2005 and June 2006; patient characteristics are shown in Table 1.

outcome measurements: surveillance BC results
A total of 773 surveillance BCs were drawn and examined. Additionally, 184 BCs (92 CVC and 92 peripheral veins) were taken in the 92 patients who developed fever in the course of neutropenia (Figure 1a). Seventy-one BCs in 45 patients were positive, 22 patients had more than one positive BC result. The identified organisms in all positive BCs are listed in Table 2. For further analysis only the first positive surveillance BC result and first diagnostic BC result were used.


Figure 1
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. (a) Flow chart per blood culture. Surveillance blood cultures were taken thrice weekly during neutropenia. Diagnostic blood cultures were taken at onset of fever. (b) Flow chart per patient.

 
Of the 92 patients with fever 42 patients were diagnosed with BSI. A summary of the clinical data for the 42 patients with fever and microbiologically documented BSI is given in Table 3.


View this table:
[in this window]
[in a new window]

  		Table 3. Summary of clinical data and BC isolates of the 42 patients with BSIs and fever

 
In 18 (43%) of those patients surveillance BC results were positive and identified microorganisms prior to onset of fever. In the remaining 24 patients BCs became positive in the samples taken at onset of fever (Figure 1b).

Three patients had positive BC results in surveillance cultures and did not present with fever as a clinical sign of BSI within the following 5 days. However, two of them were diagnosed for BSI later on. The first patient did not present with elevated temperatures: she was treated with high-dose steroids for severe graft versus host disease (GVHD) and suffered from atypical pneumonia. Stenotrophomonas maltophilia was found both in broncheo-alveolar lavage and in surveillance BCs. The second patient had temperatures up to 37.9°C and clinical deterioration. A urinary tract infection with bacteraemia and severe sepsis was diagnosed; Escherichia coli with the identical susceptibility pattern was isolated in urine cultures and in surveillance BCs. The third patient had one positive BC result with Corynebacterium sp. independently from clinical signs of infection, which was classified as ‘false positive’.

Thus, for the prediction of BSI within 5 days the PPV, NPV, sensitivity and specificity of a single positive surveillance BC was 94.7%, 70.4%, 42.9% and 98.3%, respectively. For the prediction of fever associated with a diagnostic BC consistent with the surveillance BC taken within the previous 5 days the PPV, NPV, sensitivity and specificity accounted for 85.7%, 69.7%, 42.9% and 94.8%, respectively. For the prediction of fever within 5 days irrespective of a positive diagnostic BC the PPV, NPV, sensitivity and specificity was 90%, 7.5%, 19.6% and 75%, respectively.

defervescence in patients with BSIs and positive BCs prior to onset of fever
According to the study protocol patients with detection of pathogens in surveillance BCs received specific antimicrobial therapy at onset of fever in addition to broad-spectrum antibiotics according to the standard procedure. In order to determine whether the early optimization of antimicrobial therapy according to results of positive surveillance BCs has the potential to improve clinical outcome we determined the time to defervescence in the patients with BSIs. The median time from onset of fever to defervescence was 1.5 days (range 1–6 days) and 3.5 days (range 1–18 days) in patients with BCs positive prior to onset of fever and in patients with BCs positive after onset of fever, respectively (P = 0.004, Figure 2).


Figure 2
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Identification of significant organisms causing BSIs identified by surveillance BCs allowed early optimization of antimicrobial therapy. The median time from onset of fever to devervescence was shorter in patients with BSIs and positive surveillance BCs than in patients with BSIs and BCs positive at/after onset of fever (P = 0.004).

 
patients with more than one positive BC result
Twenty-two of the 42 patients with documented BSIs had more than one positive BC. These 22 patients with BSIs and multiple positive BCs can be assigned to one of two groups: First, the 18 patients with positive surveillance BCs also had positive diagnostic BCs with the identical pathogen at/after onset of fever. Second, there were six patients with more than one positive diagnostic BC. All BC isolates (coagulase-negative staphylococci n = 5, E. coli n = 1) of these patients were resistant to the first-line antibiotic therapy with piperacillin–tazobactam, which was started at onset of fever. In these patients antibiotic therapy was adjusted once results of susceptibility testing of diagnostic BCs were available.

patients with fever but negative BCs
There were 50 patients with fever during neutropenia but negative surveillance and diagnostic BCs. This group was clinically heterogeneous as is indicated by the diagnoses: FUO, n = 31; pneumonia, n = 14 (fungal n = 6, bacterial n = 3, unknown pathogen n = 5); urinary tract infection, n = 5. Management of these patients was according to the standard clinical protocol of our institution (see ‘patients and methods’) and current guidelines. The median time from onset of fever to defervescence in these patients was 3 days (range 1–17 days), which did not differ significantly from the time to defervescence of patients with proven BSIs (median 3 days, range 1–18 days).


   discussion
Top
 Abstract
 introduction
 patients and methods
 results
 discussion
References
 
Our trial is the first prospective evaluation showing the efficacy of surveillance BCs in detecting BSIs in patients with haematologic malignancies, which are a major clinical problem [18, 19]. In the present study we were able to detect the causative organism in a relevant percentage of patients with documented BSIs prior to onset of fever by surveillance BCs.

The organisms identified prior to onset of fever were mostly Gram-positive cocci, reflecting the microbiological spectrum found in patients with antimicrobial prophylaxis for Gram-negative pathogens (here levofloxacin) [15, 20]. The majority of isolates were coagulase-negative staphylococci being indicative of systemic catheter-related infections. However, none of these patients presented with local symptoms of infection at the CVC insertion site, which is in line with previous reports that local infection is rare in catheter-related infections caused by coagulase-negative staphylococci [21]. In those patients the results of surveillance BCs enabled us to improve the clinical management: the CVCs were removed and antimicrobial therapy was optimized taking into account the results of susceptibility testing. The improved clinical management at onset of fever led to early defervescence in patients with positive surveillance BCs.

The sensitivity of surveillance BCs in detecting Gram-negative bacteraemia prior to onset of fever was low in our trial. This reflects most likely the higher virulence of these pathogens. Thus, immediate initiation of broad-spectrum antibiotic therapy, independent of surveillance BC results, at onset of neutropenic fever is mandatory should this new strategy be introduced into standard care.

A common concern of the procedure of taking surveillance BCs are false-positive results and a potential ‘over-treatment’ leading to drug-related side-effects and high costs [22–24]. In our study we investigated exclusively highly immunosuppressed patients with prolonged neutropenia and found a low rate of false-positive surveillance BC results. We certainly can not exclude that some of the surveillance BC results detecting coagulase-negative staphylococci were rather a sign of catheter colonization than of significant infection. However, since we found that catheter removal and start of specific antimicrobial therapy led to a shorter period of fever in patients in the surveillance BC group we consider the majority of positive surveillance BC results to be ‘true positive’.

Our results highlight that prospective surveillance BCs are specific and have a high predictive value for BSIs in neutropenic patients, supporting their use in guiding treatment. However, their sensitivity and negative predictive value for fever in neutropenia is low. Thus, it remains to be determined whether this approach is a cost-effective use of resources.

The impact of surveillance BCs on the outcome in neutropenic patients with BSIs has to be assessed in a randomized controlled trial. The outline design for such a study could be: randomization of patients with positive surveillance BCs to receive either (i) empirical antibiotic therapy (standard) or (ii) empirical antibiotics plus directed antimicrobial therapy according to the pathogen isolated in the surveillance BC (intervention). The key outcome measure would be the time to devervescence. Secondary outcome measurements should include survival and a cost–benefit analysis.

Received for publication March 28, 2007. Revision received June 6, 2007. Accepted for publication June 8, 2007.


   References
Top
 Abstract
 introduction
 patients and methods
 results
 discussion
 References
 
1. Almyroudis NG, Fuller A, Jakubowski A, et al. Pre- and post-engraftment bloodstream infection rates and associated mortality in allogeneic hematopoietic stem cell transplant recipients. Transpl Infect Dis (2005) 7:11–17.[CrossRef][Web of Science][Medline]

2. Bearman GM, Wenzel RP. Bacteremias: a leading cause of death. Arch Med Res (2005) 36:646–659.[CrossRef][Web of Science][Medline]

3. Guven GS, Uzun O, Cakir B, et al. Infectious complications in patients with hematological malignancies consulted by the Infectious Diseases team: a retrospective cohort study (1997–2001). Support Care Cancer (2006) 14:52–55.[CrossRef][Web of Science][Medline]

4. Penack O, Beinert T, Buchheidt D, et al. Management of sepsis in neutropenia: guidelines of the infectious diseases working party (AGIHO) of the German Society of Hematology and Oncology (DGHO). Ann Hematol (2006) 85:424–433.[CrossRef][Web of Science][Medline]

5. Mermel LA, Farr BM, Sherertz RJ, et al. J Intraven Nurs (2001) 24:180–205.[Medline]

6. Hughes WT. Bloodstream infections in cancer patients. Eur J Cancer (2005) 41:1370–1371.[CrossRef][Web of Science][Medline]

7. Coullioud D, Van der AP, Viot M, Lasset C. Prospective multicentric study of the etiology of 1051 bacteremic episodes in 782 cancer patients. CEMIC (French-Belgian Study Club of Infectious Diseases in Cancer). Support Care Cancer (1993) 1:34–46.[CrossRef][Web of Science][Medline]

8. Hughes WT, Armstrong D, Bodey GP, et al. 2002 guidelines for the use of antimicrobial agents in neutropenic patients with cancer. Clin Infect Dis (2002) 34:730–751.[CrossRef][Web of Science][Medline]

9. Link H, Bohme A, Cornely OA, et al. Antimicrobial therapy of unexplained fever in neutropenic patients–guidelines of the Infectious Diseases Working Party (AGIHO) of the German Society of Hematology and Oncology (DGHO), Study Group Interventional Therapy of Unexplained Fever, Arbeitsgemeinschaft Supportivmassnahmen in der Onkologie (ASO) of the Deutsche Krebsgesellschaft(DKG-German Cancer Society). Ann Hematol (2003) 82:S105–117.[CrossRef][Web of Science][Medline]

10. Kirby JT, Fritsche TR, Jones RN. Influence of patient age on the frequency of occurrence and antimicrobial resistance patterns of isolates from hematology/oncology patients: report from the Chemotherapy Alliance for Neutropenics and the Control of Emerging Resistance ProGram (North America). Diagn Microbiol Infect Dis (2006) 56:75–82.[CrossRef][Web of Science][Medline]

11. Leibovici L, Paul M, Cullen M, et al. Antibiotic prophylaxis in neutropenic patients: new evidence, practical decisions. Cancer (2006) 107:1743–1751.[CrossRef][Web of Science][Medline]

12. Micozzi A, Bucaneve G. Prophylaxis and treatment of bacterial infections: do we need new strategies? Rev Clin Exp Hematol (2005) 9:20–23.

13. Neuburger S, Maschmeyer G. Update on management of infections in cancer and stem cell transplant patients. Ann Hematol (2006) 85:345–356.[CrossRef][Web of Science][Medline]

14. Penack O, Keilholz U, Thiel E, Blau IW. Value of surveillance blood cultures in neutropenic patients–a pilot study. Jpn J Infect Dis (2005) 58:171–173.[Medline]

15. Chizuka A, Kami M, Kanda Y, et al. Value of surveillance blood culture for early diagnosis of occult bacteremia in patients on corticosteroid therapy following allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant (2005) 35:577–582.[CrossRef][Web of Science][Medline]

16. Carlisle PS, Gucalp R, Wiernik PH. Nosocomial infections in neutropenic cancer patients. Infect Control Hosp Epidemiol (1993) 14:320–324.[Web of Science][Medline]

17. Gaynes RP, Horan T. Surveillance of nosocomial infections. In: CDC definitions of nosocomial infections (2004) Lippincott Williams and Wilkins. 1659–1702.

18. Dettenkofer M, Wenzler-Rottele S, Babikir R, et al. Surveillance of nosocomial sepsis and pneumonia in patients with a bone marrow or peripheral blood stem cell transplant: a multicenter project. Clin Infect Dis (2005) 40:926–931.[CrossRef][Web of Science][Medline]

19. Wisplinghoff H, Cornely OA, Moser S, et al. Outcomes of nosocomial bloodstream infections in adult neutropenic patients: a prospective cohort and matched case-control study. Infect Control Hosp Epidemiol (2003) 24:905–911.[CrossRef][Web of Science][Medline]

20. Serody JS, Berrey MM, Albritton K, et al. Utility of obtaining blood cultures in febrile neutropenic patients undergoing bone marrow transplantation. Bone Marrow Transplant (2000) 26:533–538.[CrossRef][Web of Science][Medline]

21. Safdar N, Maki DG. Inflammation at the insertion site is not predictive of catheter-related bloodstream infection with short-term, noncuffed central venous catheters. Crit Care Med (2002) 30:2632–2635.[CrossRef][Web of Science][Medline]

22. Barton T, Mizuta M, Fishman N. Costs of false-positive cultures from central catheters. Ann Intern Med (2004) 141:162.[Free Full Text]

23. Bates DW, Goldman L, Lee TH. Contaminant blood cultures and resource utilization. The true consequences of false-positive results. JAMA (1991) 265:365–369.[Abstract/Free Full Text]

24. Souvenir D, Anderson DE Jr, Palpant S, et al. Blood cultures positive for coagulase-negative staphylococci: antisepsis, pseudobacteremia, and therapy of patients. J Clin Microbiol (1998) 36:1923–1926.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Ann OncolHome page
O. Penack, P. Rempf, B. Graf, I. W. Blau, and E. Thiel
Aspergillus galactomannan testing in patients with long-term neutropenia: implications for clinical management
Ann. Onc., May 1, 2008; 19(5): 984 - 989.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/11/1870    most recent
mdm351v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Penack, O
Right arrow Articles by Blau, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Penack, O
Right arrow Articles by Blau, I.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?