A Spoonful of Sugar Helps the Negative Blood Culture Rate Down
Abstract & Commentary
Synopsis: The number of episodes of bacteraemia detected was increased by one-third simply by inoculating blood into blood culture medium supplemented with 5% sucrose, creating a hypertonic environment with which cell-wall deficient bacteria could grow.
Source: Woo PC, et al. Lancet. 2001;357:675-679.
About 4 of 5 episodes of neutropenic fevers are thought to be caused by infections, but an isolate is detected in fewer than half of these cases. One explanation might be that some of the bacteria involved might possess a deficient cell wall because of exposure to antibiotics—particularly the beta-lactam antibiotics. Woo and associates set out to investigate the role of cell-wall-deficient bacteria, if any, in causing infection in recipients of an haematopoietic stem cell transplant (HSCT). Recipients’ blood cultures were obtained when patients developed fever and were inoculated into an aerobic bottle with resin, an anaerobic bottle, and a hypertonic bottle containing 5% sucrose to isolate cell-wall deficient bacteria. When growth was detected in this bottle, a small sample of the broth was subcultured onto solid media, and the isolate was identified by standard biochemical methods. While 55 episodes of bacteraemia due to normal bacteria were detected in 15 (17%) of 86 bone marrow transplant (BMT) recipients enrolled into the study, a further 20 episodes due to cell-wall deficient bacteria were detected in 12 (14%) patients. Antibiotic treatment was successful in 19 (95%) of these episodes. A beta-lactam antibiotic, a glycopeptide, or both had been administered in 16 cases within 10 days before blood cultures had been taken. The majority of isolates detected during neutropenia were Gram-positive bacteria—half of which were cell-wall deficient and only detected in the sucrose-enriched media (see Table).
| Table-Table of Isolates | |||
| Gram reaction | Isolate | Cell-wall deficient |
Normal bacteria |
| Gram-positive | 11 | 11 | |
| Cocci | 3 | 9 | |
| Staphylococcus species | 1 | 4 | |
| Streptococcus species | 1 | 3 | |
| Micrococcus species | 1 | 0 | |
| Enterococcus species | 0 | 2 | |
| Bacilli | 8 | 2 | |
| Bacillus species | 7 | 2 | |
| Lactobacillus species | 1 | 0 | |
| Gram-negative bacilli | 2 | 6 | |
| Enterobacteriaceae | 0 | 2 | |
| Non-fermentors | 2 | 4 | |
| Total | 13 | 17 | |
These results indicate that bacteraemia due to cell-wall-deficient bacteria causes a significant proportion of so-called culture-negative febrile episodes in HSCT recipients.
Comment by J. Peter Donnelly, Phd
Cell-wall deficient bacteria or L-forms are known to be induced by both in vitro and in vivo antibiotics that act specifically on cell-wall synthesis. These include the beta-lactam antibiotics, penicillins, cephalosporins, and carbapenems, as well as the glycopeptides, vancomycin and teicoplanin. That the majority of cell-wall deficient species were Gram-positive bacteria is also not a surprise since their cell wall is exposed to antibiotic action consisting as it does of a thick peptidoglycan layer. In contrast, the cell wall of the Gram-negative bacilli such as Escherichia coli, Pseudomonas aeruginosa and nonfermenting bacilli is enveloped by an outer membrane consisting of lipopolysaccharide (endotoxin), which acts as both a barrier and an osmotic stabilizer making the formation of cell-wall deficient forms less likely. Neither is prior exposure to cell-wall active antibiotics an unusual finding. However, the results of this study pose a few intriguing questions.
First, if the bacteraemia rate can be almost doubled by simply including an extra bottle of broth with some sugar, why isn’t everyone doing it? In fact, this is an old chestnut that has been roasted before. Twenty-five years ago, Washington and colleagues already investigated the value of using hypertonic medium for detecting bacteraemia and concluded "The addition of sucrose . . . did not significantly increase the rate of positivity or the time interval to detection of positivity of any group of bacteria."1 Later, others came to a different conclusion reporting that hypertonic medium offers no advantage in the recovery of anaerobes but is of value in the recovery of facultative anaerobes (ie, the bacteria in question).2 Later, another group demonstrated that the growth of both Staphylococcus aureus and Enterobacteriaceae occurred earlier or solely in hypertonic broth cultures collected during treatment of patients with beta-lactam antibiotics.3 This was confirmed by another study in which the recovery not only of Gram-positive and Gram-negative facultatively anaerobic bacteria was improved using hypertonic broth but also that of yeasts.4 Since then, the field has lain fallow.
Second, will it really make a difference, besides adding to the costs, if all haematology departments were to adopt a hypertonic medium? Frankly, I doubt it. Most, like Woo et al, already use broad-spectrum antibiotics, which, though not ideal for treating Gram-positive bacterial infections, do deal remarkably effectively with the vast majority of neutropenic fevers resulting in few deaths due to bacteria even without the addition of a glycopeptide—a fact that Woo et al readily admit.
Last, are these cell-wall deficient bacteria any other than an interesting curiosity? Probably not. Detecting them can be interpreted as either a sign that the antibiotic regimen is failing or as an indication that although not optimum, the drugs are affecting the bacteria adversely in damaging their cell wall, rendering such cells osmostically unstable. Only if they were to cause excess mortality or lead to relapse need one draw any great significance from recovering them from blood cultures since the organisms involved were almost exclusively the indolent bacterial species that are least associated with complications. The only exception in the study reported was the case with bacteraemia due to the enteric bacillus Serratia marcescens, which were recovered only in a hypertonic medium and, despite being susceptible to the antibiotic regimen used for treatment, nonetheless went on to disseminate, fulminate, and result in death. This case defied explanation microbiologically and unfortunately probably would have proven fatal whichever medium was used to detect it.
References
1. Washington JA 2nd, et al. J Clin Microbiol. 1975;1: 79-81.
2. Ellner PD, et al. J Clin Microbiol. 1976;4:216-224.
3. Eng J, Maeland A. J Clin Microbiol. 1982;16:890-894.
4. Reimer LG, et al. J Clin Microbiol. 1983;17:1045-1049.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content