Journal of Antimicrobial Chemotherapy (2000) 45, 123-127
© 2000 The British Society for Antimicrobial Chemotherapy
Brief reports |
Serum bactericidal and inhibitory titres in the management of melioidosis
a Faculty of Tropical Medicine, Mahidol University, 420/6 Rajvithi Road, Bangkok 10400, Thailand; b Centre for Tropical Medicine, Nuffield Department of Clinical Medicine, John Radcliffe Hospital, University of Oxford, Oxford, UK
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsDiscussionReferences |
|---|
A retrospective evaluation of the relationship between serum bactericidal and inhibitory titres and treatment outcome in 195 adult Thai patients with severe melioidosis was conducted. Drug regimens included ceftazidime (52% of patients), co-amoxiclav (24%), imipenem (11%) or the conventional four-drug combination (11%). Pre- and 1 h post-dose serum samples were collected after 48–72 h of therapy, and serum inhibitory and bactericidal titrations determined. Median post-dose titres were: bactericidal 1:8 (range 0–1:128) and inhibitory 1:16 (range 0–1:128). Overall mortality was 26% and outcome was not influenced by either inhibitory or bactericidal titres. Pre-dose titres correlated with renal function; renal function was the most important predictor of mortality. Determination of serum inhibitory or bactericidal titres is unhelpful in the management of severe melioidosis.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Melioidosis, caused by Burkholderia pseudomallei, is a common cause of community-acquired septicaemia in endemic areas, such as north-east Thailand.1 The majority of patients are bacteraemic on admission to hospital.2 Conventional four-drug therapy (a combination of iv chloramphenicol, doxycycline and trimethoprim– sulphamethoxazole) was associated with a mortality in septicaemic melioidosis of c. 80%. Ceftazidime has reduced mortality considerably, but not below c. 40% in severe disease. Imipenem and co-amoxiclav are effective alternatives.3,4 Recovery is usually slow and intravenous therapy needs to be given for at least 10–14 days. Prolonged fever or continued positive blood cultures are common despite adequate dosing, as is subsequent relapse. More effective therapeutic regimens are required and there is a lack of useful tests to guide therapy.
We have evaluated retrospectively the prognostic value of serum bactericidal and inhibitory titres in determining treatment outcome in severe melioidosis.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Clinical procedures
Adult patients (over 14 years old), admitted to Sappasitprasong Hospital with suspected severe melioidosis, were enrolled into one of a series of treatment trials, as described previously, if they or their relatives gave informed consent.2–4 Patients were randomized to either ceftazidime or one of intravenous ‘conventional’; four-drug therapy, co-amoxiclav or imipenem. The drug regimens consisted of: ceftazidime 120 mg/kg/day in three divided doses by slow iv injection; chloramphenicol succinate 100 mg/kg/ day, doxycycline 4 mg/kg/day, trimethoprim 10 mg/kg/day and sulphamethoxazole 50 mg/kg/day in divided doses; co-amoxiclav 160 mg/kg/day in six divided doses by slow iv injection; or imipenem 50 mg/kg/day in three divided doses by iv infusion over 90 min. Pre-dose (immediately before) and post-dose (1 h after slow iv injection dose or end of the infusion) samples were collected, 48–72 h after starting therapy, by venepuncture. Specimens were centrifuged as soon as possible at 3000 rpm at 4°C. Serum was stored at –30°C until assays could be performed, if these could not be performed immediately.
Laboratory procedures
B. pseudomallei was isolated as described previously.5 Paired serum specimens from each patient were serially double-diluted in tryptic soy broth (TSB) (Unipath, Basingstoke, UK) in U-well microtitre plates to give a range of titres of 1:2–1:128. The patient's original isolate of B. pseudomallei was incubated overnight in TSB and then diluted to achieve a final inoculum per well of approximately 1 x 105 cfu/mL. Precise inocula were determined by dilution and sub-culture. Plates were incubated overnight at 35–37°C in air, unshaken. The highest dilution without visible growth was recorded as the serum inhibitory titre (SIT) and 20 µL volumes from each well were subcultured on to Columbia agar. After overnight incubation at 37°C, plate colony counts were performed. The highest dilution at which 
99.9% kill of the original inoculum was achieved was recorded as the serum bactericidal titre (SBT). Titres from some of these patients have been published previously.3,5
Serum creatinine concentrations in samples collected from patients on admission were determined in the Biochemistry Laboratory of Sappasitprasong Hospital using an automated alkaline picrate method. Data analysis was performed using the statistical computing package SPSS 8.0 for Windows (SPSS Inc., Chicago, IL, USA).
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
Serum bactericidal titre results were available for 195 melioidosis patients and inhibitory titres for 191 patients. There were 129 male patients (66.2%). The median age was 50 years (range 16–78). Drug therapy was as follows: ceftazidime, 100 patients; co-amoxiclav, 48; imipenem, 22; and four-drug combination, 21 patients. Mortality was 22, 25, 23 and 52%, respectively. Patients treated with the four-drug regimen had a higher mortality than those treated with the ß-lactams (mortality 52% versus 23%, P = 0.008). Four patients received other regimens (doxycycline plus chloramphenicol, ceftazidime plus oral ciprofloxacin, ciprofloxacin alone and ampicillin–sulbactam). Overall mortality was 26%; 116 patients (59.5%) were bacteraemic at admission and 41 of these died (35.3%).
Titre results for the four main drug regimens are shown in Table I
. There were no significant differences overall between sexes for SIT or SBT and no significant associations with age. There were strong positive correlations overall between pre- and post-dose titres for both inhibitory and bactericidal determinations overall, as well as between inhibitory and bactericidal titres either pre- or post-dose (all correlations, Spearman rank correlation coefficient r > 0.74, P < 0.001).
|
Outcome
The post-dose bactericidal titres for all patients, by outcome, are shown in Figure 1. Overall, higher inhibitory or bactericidal titres were not associated with improved survival, either pre- or post-dose. However, for ceftazidime, higher post-dose SBTs were associated significantly with mortality (survivors, median = 16, interquartile (IQ) range 8–32; non-survivors, median = 32, IQ range 16–64, P = 0.015), as were higher pre-dose SBTs (median 4 versus 12, respectively, P = 0.003) and SITs (median 8 versus 24, P = 0.022). For imipenem, pre-dose bactericidal titres only were associated with higher mortality (P = 0.03). Higher titres for co-amoxiclav and the four-drug regimen were not significantly associated with death.
|
Among bacteraemic patients, there were no significant associations overall between SBT or SIT and outcome. For ceftazidime-treated patients only, there were significant associations with increased mortality for higher SBTs, both pre-dose (survivors, median = 8; non-survivors, median = 24, P = 0.047) and post-dose (medians = 16 versus 32, respectively, P = 0.024).
SBT/SIT ratios
Both overall, and for the individual drug regimens, higher pre- and post-dose SBT/SIT ratios were not associated with improved outcome.
Post-dose/pre-dose SIT and SBT ratios
The ratios of post- to pre-dose titres (both inhibitory and bactericidal) are shown in Figure 2. Overall there was an association between SIT post-/pre-dose ratios and outcome (survivors, median ratio 2, IQ range 2–4; non-survivors, median ratio 2, IQ range 1–4, P = 0.032), but not between SBT ratios and outcome (P = 0.525). There were no associations between outcome and ratios for any of the individual drug regimens, or the three ß-lactam agents combined.
|
Renal function
Admission renal function was correlated positively with mortality overall, i.e. patients with impaired renal function were more likely to die. Admission serum creatinine concentrations and related mortality are shown in Table II.
|
Among those patients with normal renal function (serum creatinine concentrations
1.7 mg/dL; n = 96), there were no associations between outcome and titres (P > 0.5 for all titres), or SIT post-/pre-dose ratios (P = 0.91). This was also true of patients in the lowest quartile for serum creatinine concentrations (i.e. < 1.2 mg/dL). Overall, there were positive correlations between serum creatinine concentrations and pre-dose titres (SIT, r = 0.194, P = 0.009; SBT, r = 0.20, P = 0.006), but not post-dose titres. Creatinine concentration was correlated with pre-dose titres (SIT, r = 0.38, P = < 0.001; SBT, r = 0.37, P = < 0.001) for ceftazidime-treated patients, and with pre-dose SBT for imipenem (r = 0.54, P = 0.009), but not with titres for either co-amoxiclav or the four-drug regimen.
Those patients who had a pre-dose bactericidal titre 1:4 had a mean creatinine of 2.47 mg/dL (95% CI 2.10–2.90) but for patients with a titre < 1:4 the mean creatinine was 1.85 mg/dL (1.63–2.09), P = 0.005. This was also true of pre-dose inhibitory titres (means 2.32 versus 1.87 mg/dL, respectively, P = 0.04). Post-dose titres of > 1:4 were not significantly associated with renal function.
Logistic regression analyses were performed to examine the relative influence on mortality of renal function and both SITs and SBTs. These multivariate analyses compared age, gender, drug regimen, sGOT, serum creatinine concentration, plasma lactate, the presence of positive blood cultures and either SIT or SBT (pre-dose titres > 1:4). Serum creatinine (SBT, P = 0.006, OR 1.32, 95% CI 1.08–1.60) and plasma lactate (SBT, P = 0.11, OR 1.18, 95% CI 0.96–1.46) concentrations were the most important predictors of mortality. Similar results were found for inhibitory titres.
There were significant inverse correlations between serum creatinine concentration and either bactericidal (r = 0.29, P = 0.006) or inhibitory (r = -0.31, P = <0.001) post-/pre-dose ratios. In a multiple logistic regression analysis as described above, renal function was the most important predictor of mortality (P = 0.004, OR 1.37, 95% CI 1.11–1.71) and neither SIT nor SBT ratios were independent risk factors.
Duration of hospital stay and fever clearance times
The median duration of hospital stay in survivors was 22 days (range 4–66). Both overall, and for patients with normal renal function, there were no significant correlations between SIT, SBT or post-/pre-dose ratios, and either duration of hospital stay or fever clearance. There was no evidence to suggest that those surviving patients with the lowest SBTs or SITS ( 1:2) had longer stays in hospital or prolonged fever.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
The value of serum bactericidal and inhibitory titres in guiding antibiotic therapy or as prognostic indicators, for any infection, remains in doubt.6,7 There is little clinical evidence to suggest a benefit, but the few reported studies have involved small numbers of patients. Only one study in bacteraemic patients has shown a statistically significant association between outcome and titres (
1:8).8 The SBT test is most commonly used for subacute infections such as infectious endocarditis (very rare in melioidosis), but there is debate over the value of SBT determination even in this disease.9 There has also been considerable variation in the application and interpretation of the test.10 Our results demonstrate that SBT and SIT determinations are not useful in the individual management of severe melioidosis. Outcome, duration of stay in hospital and fever clearance times were all unrelated to these titres. The three ß-lactam antibiotics all achieved reasonable inhibitory and bactericidal activity in serum, although the four-drug regimen had relatively poor inhibitory activity and very little bactericidal action. This is reflected in the considerably higher mortality in this group of patients.
In the present study, significantly higher inhibitory and bactericidal titres, particularly pre-dose, were seen in patients who subsequently died. These higher pre-dose titres were a result of the greater degree of renal impairment present in those patients who died; such impairment of renal function was the most important determinant of mortality. Although high ratios between post- and pre-dose titres appeared to be beneficial, this effect was lost when renal function was taken into account. This relationship between renal impairment and higher SITs and SBTs is unsurprising, as such functional impairment reduces the urinary excretion of antibiotics, the main route of elimination of the ß-lactam agents. Furthermore, the principal determinant of therapeutic outcome when severe bacterial infections are treated with ß-lactam antibiotics is the time (T) for which plasma concentrations remain above the MIC for the infecting organism. The steep concentration–effect relationship with these drugs and their usually rapid elimination result in an abrupt transition from maximum activity (Emax) to no activity, once levels have fallen to the MIC. The peak and trough titres measured here are an imprecise measure of T. The patients with renal impairment had extremely high levels, probably achieving Emax throughout the dosing interval. Relationships between SIT or SBT and therapeutic effect would therefore not have been expected in this group, but nor were we able to demonstrate any such relationship in patients with normal renal function.
In conclusion, determination of serum bactericidal activity does not provide useful prognostic information and is of no value in the management of acute severe melioidosis.
|
Acknowledgments |
|---|
We are grateful to Julie Simpson for statistical advice and to Paul Howe for assistance in performing assays. We thank the Director of Sappasitprasong Hospital, Ubon Ratchathani, and Drs Wipada Chaowagul, Adul Rajanuwong and the medical and nursing staff of the Department of Medicine for their help with our studies. Drs Yupin Suputtamongkol, Mike Smith and Brian Angus helped with collection of specimens. This study was part of the Wellcome–Mahidol University, Oxford Tropical Medicine Research Programme, funded by the Wellcome Trust of Great Britain.
|
Notes |
|---|
* Corresponding author. Tel: +66-2-246-0832; Fax: +66-2-246-7795; E-mail: fnajs{at}diamond.mahidol.ac.th
Present address. Public Health Laboratory, Derriford Hospital, Plymouth, UK
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionReferences |
|---|
1 . Chaowagul, W., White, N. J., Dance, D. A., Wattanagoon, Y., Naigowit, P., Davis, T. M. et al. (1989). Melioidosis: a major cause of community-acquired septicemia in northeastern Thailand. Journal of Infectious Diseases 159, 890–9.[Web of Science][Medline]
2 . White, N. J., Dance, D. A., Chaowagul, W., Wattanagoon, Y., Wuthiekanun, V. & Pitakwatchara, N. (1989). Halving of mortality of severe melioidosis by ceftazidime. Lancet ii, 697–701.
3 . Simpson, A. J., Suputtamongkol, Y., Smith, M. D., Angus, B. J., Rajanuwong, A., Wuthiekanun, V. et al. (1999). Comparison of imipenem and ceftazidime as therapy for severe melioidosis. Clinical Infectious Diseases 29, 381–7.[Web of Science][Medline]
4 . Suputtamongkol, Y., Rajchanuwong, A., Chaowagul, W., Dance, D. A., Smith, M. D., Wuthiekanun, V. et al. (1994). Ceftazidime vs. amoxicillin/clavulanate in the treatment of severe melioidosis. Clinical Infectious Diseases 19, 846–53.[Web of Science][Medline]
5 . Walsh, A. L. & Wuthiekanun, V. (1996). The laboratory diagnosis of melioidosis. British Journal of Biomedical Science 53, 249–53.[Web of Science][Medline]
6 . Wolfson, J. S. & Swartz, M. N. (1985). Drug therapy. Serum bactericidal activity as a monitor of antibiotic therapy. New England Journal of Medicine 312, 968–75.[Web of Science][Medline]
7 . Reller, L. B. (1986). The serum bactericidal test. Reviews of Infectious Diseases 8, 803–8.[Web of Science][Medline]
8 . Klastersky, J., Munier-Carpentier, F. & Prevost, J.-M. (1977). Significance of antimicrobial synergism for the outcome of gram negative sepsis. American Journal of Medical Science 273, 157–67.
9
.
Working Party of the British Society for Antimicrobial Chemotherapy. (1998). Antibiotic treatment of streptococcal, enterococcal and staphylococcal endocarditis. Heart 79, 207–10.
10
.
MacGowan, A., McMullin, C., James, P., Bowker, K., Reeves, D. & White, L. (1997). External quality assessment of the serum bactericidal test: results of a methodology/interpretation questionnaire. Journal of Antimicrobial Chemotherapy 39, 277–84.
Received 22 February 1999; returned 12 May 1999; revised 24 June 1999; accepted 31 August 1999
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. C. Cheng and B. J. Currie Melioidosis: Epidemiology, Pathophysiology, and Management Clin. Microbiol. Rev., April 1, 2005; 18(2): 383 - 416. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||