Objective To investigate current practices with teicoplanin in patients with haematological malignancy in centres throughout the UK and Ireland with respect to indication for usage and timing of introduction in febrile neutropenia, dosage regimens and therapeutic drug monitoring practices.
Methods An online survey was distributed to 598 haematology and oncology pharmacists representing 168 institutions throughout the UK and Ireland. Survey questions were aimed at identifying typical hospital practices for teicoplanin use specifically in patients with haematological malignancy in terms of empiric use strategies, dosage regimens and therapeutic drug monitoring. Participants were asked to base their answers on actual practice or policy in their hospitals and not on personal opinions.
Results A total of 51 pharmacists participated in the survey. Responses indicated that teicoplanin is widely used in adult patients with haematological malignancy in the UK and Ireland, but evidence–practice gaps for empiric use strategies in febrile neutropenia were noted. For dose selection, the manufacturer's Summary of Product Characteristics was heavily relied upon in UK and Irish institutions, rather than therapeutic drug monitoring, as an indicator of therapeutic dosing.
Conclusions Despite emerging evidence to support targeted prescribing, aggressive dosing and routine therapeutic drug monitoring, findings suggest that many centres do not use teicoplanin in this way. The findings suggest inappropriate use of teicoplanin in these patients. Improving the translation of available evidence into regular practice may improve patient outcomes and reduce unnecessary teicoplanin usage. Pharmacists could aid this process through education and increased involvement in drug therapy decisions.
- Haematological Malignancy
- Febrile Neutropenia
Statistics from Altmetric.com
The glycopeptide antibiotic teicoplanin represents a widely available and possibly safer alternative to vancomycin for methicillin-resistant Gram-positive infection. Comparative studies versus vancomycin have shown teicoplanin to be equally effective but with a lower incidence of nephrotoxicity.1
Current ‘best practice’ for teicoplanin use in febrile neutropenia is well documented in international recommendations and guidelines. This practice involves restricting the application of teicoplanin to only those clinical situations where substantial benefit is likely to be achieved from such therapy. Persistent fever alone, in a clinically stable patient, is no longer considered to be an indication for the empiric addition of a glycopeptide.2–4 This approach resulted from resistance concerns, especially among Enterococci, and the publication of several important studies showing that the routine addition of a glycopeptide in the setting of persistent fever before documentation of a Gram-positive infection does not improve outcome.5–7 There are no national guidelines for the management of febrile neutropenia in Ireland, and there were none in the UK until the publication of National Institute for Health and Care Excellence (NICE) Guidelines in September 2012,4 so decisions have generally been made locally.8 Considerable variation in practices between UK haematology units with regard to the antibiotic management of febrile neutropenia has been previously reported.9 The current situation in terms of teicoplanin use in febrile neutropenia is unclear however, and anecdotally, the optimal time to start teicoplanin and the appropriate indications for its use may be controversial among clinicians.
Clinical studies continue to question whether the current dosage recommendation for teicoplanin in the manufacturer's Summary of Product Characteristics (SmPC) is adequate for the treatment of infections in patients with haematological malignancy.10–12 The current SmPC recommends 3×400 mg 12 hourly, then 400 mg once daily for most severe infections, including complicated skin and soft tissue infections, complicated urinary tract infections and pneumonia. Higher doses are recommended for bone and joint infections and infective endocarditis.13 Altered pharmacokinetic behaviour of teicoplanin, such as a larger volume of distribution and increased renal clearance, has been observed in febrile neutropenic patients with haematological malignancy.10 These pharmacokinetic changes may result in lower than expected serum concentrations of teicoplanin and are important considerations for appropriate dosing. The underlying mechanism for these changes is unclear, but factors such as hypoalbuminaemia, increased fluid load, capillary leakage, augmented renal clearance (ARC) and the malignancy itself may be involved.11 ,14
Teicoplanin has a wide therapeutic window and low potential for toxicity.15 Therapeutic drug monitoring (TDM) is recommended to optimise therapy, rather than simply to avoid toxicity, because a relationship between serum concentration and toxicity has not been established.13 Current therapeutic serum trough (predose) concentrations are defined as >15 mg/L for most infections,13 but a higher minimum threshold of >20 mg/L is suggested for some serious infections such as endocarditis, osteomyelitis and septic arthritis,15 and in immunocompromised patients.11 It is generally recommended to keep trough levels below 60 mg/L to avoid toxicity, but there is a lack of evidence to support this concern.16
At present there is no specific dosing or monitoring practice recommended for teicoplanin use when treating patients with haematological malignancy. Centres therefore have to make their own decisions on what might best serve this patient group, and anecdotally, some have opted to use a specific protocol for these patients. Important considerations are that indwelling intravascular access is common, and the risk of serious Gram-positive infection is high. Moreover, these patients are often severely neutropenic, and rapid and effective treatment of infection where it occurs is vital.
The present study aimed to identify the current situation with teicoplanin use in patients with haematological malignancy in UK and Irish institutions with respect to indications for usage and timing of introduction in febrile neutropenia, dosage and TDM practices.
The survey was constructed using the electronic SurveyMonkey tool (http://www.surveymonkey.com). Survey questions were aimed at identifying typical hospital practices for teicoplanin use in patients with haematological malignancy in terms of empiric use strategies, dosage regimens and TDM. Questions relating to the empiric use of teicoplanin in febrile neutropenia were based on the Infectious Diseases Society of America (IDSA) 2010 update of the Clinical Practice Guideline for use of Antimicrobials in Neutropenic Patients with Cancer.3 Basic hospital demographic information was requested without identifying respondents or their institutions. The survey was piloted within a single institution on a total of seven personnel: two haematologists and five clinical pharmacists. The ‘thinking aloud’ method was used to provide feedback for refinement of questions to ensure validity and reliability of responses.17
Survey participants and distribution
A formal application to the British Oncology Pharmacy Association (BOPA) to survey its members was accepted in November 2012. The survey was distributed to 598 BOPA members from 168 institutions throughout the UK and Ireland. Each member was sent an email with an invitation to complete an electronic questionnaire-based anonymous survey. A hardcopy option of the questionnaire and an explanatory letter were attached to this email. Participants were asked to refine answers to practice in patients with haematological malignancy rather than other oncology patients and to base their answers on actual practice or policy in their hospitals and not on personal opinions. The survey was made available from 6 December 2012 until 25 January 2013. Two reminder emails were sent on 17 December 2012 and 9 January 2013.
Survey responses were imported into an IBM SPSS Statistics V.19 database for quantitative statistical analysis.
A total of 51 responses were received: 45 of the respondents (88%) used teicoplanin to treat infection in adult patients with haematological malignancy. In the remaining six responses (12%), teicoplanin was not used at all for this patient group and these responses were excluded from further analyses. The practice settings for respondents using teicoplanin were district general hospitals (27/45, 60%), tertiary referral hospitals (12/45, 27%), cancer centres (4/45, 9%) and private hospitals (2/45, 4%). No significant differences were found between institution types in terms of empiric use strategies in febrile neutropenia, dosage regimens or TDM.
Empiric use strategies in febrile neutropenia
Fifty-seven per cent (27/47) of respondents use teicoplanin empirically in patients with febrile neutropenia. Responses revealed considerable variation in approach to the timing of teicoplanin introduction to empiric therapy in febrile neutropenia (figure 1). The most common approach for the time that teicoplanin is normally added to therapy was second line when fever persisted, either with or without changing initial antibiotics (13/27, 48%). Current ‘best practice’ for targeted use of teicoplanin in febrile neutropenia was reported in 10 responses (37%).
Clinical situations selected by respondents (14) that might prompt the inclusion of teicoplanin in the initial empiric antibiotic regimen for febrile neutropenia, when this was not routine practice, are shown in figure 2.
The manufacturer's SmPC dosage recommendation for teicoplanin in severe infection (3×400 mg 12 hourly, then 400 mg once daily) was used by the majority of respondents for empiric use in febrile neutropenia (18/25, 72%) and for documented infection (34/44, 77%) in patients with haematological malignancy and normal renal function.13 All respondents reported using the same dosage regimen regardless of the underlying malignancy. When teicoplanin was initiated empirically, most respondents would not change the regimen if a particular infection was subsequently documented (19/24, 79%).
Those respondents not following the manufacturer's SmPC dosage recommendation used a range of dosing regimens, but all used higher doses. The main source of guidance for dosing in these centres was a local clinical decision (6/11, 55%). Other regimens adopted included a single 1200 mg loading dose followed by 800 mg daily; 3×800 mg 12 hourly, then 800 mg daily; and 3×12 mg/kg 12 hourly, then 12 mg/kg daily.
Therapeutic drug monitoring
Only 12% of respondents (5/42) reported that they conduct TDM routinely during teicoplanin therapy in patients with haematological malignancy. When these respondents were asked to state the minimum serum trough concentration routinely targeted in this patient group in their hospital, two specified 20 mg/L, one specified 10 mg/L and two were ‘not sure’. Microbiological advice was the only reason listed for changing the target trough concentration in a particular type of infection (2/5, 40%).
The survey findings suggest that teicoplanin is widely used in adult patients with haematological malignancy in the UK and Ireland, but there is considerable variation in the approach to using this drug. Moreover, in some instances, there is a marked difference between what has been recommended by international guideline groups and local practice. This difference between guidelines and practice is reminiscent of similarly described examples across many different treatment modalities, and is not confined to this clinical issue. Indeed, it is reported to take an average of 17 years to translate new knowledge from clinical trials into practice, and even then application is highly variable.18 The closing of this knowledge–practice gap has the potential to benefit patient care significantly.
The empiric use of teicoplanin in the setting of persistent fever still appears to be common in haematology units in the UK and Ireland despite evidence of a lack of benefit in terms of mortality or reduction in time to defervescence,5–7 and despite published guidelines promoting targeted use of glycopeptides in febrile neutropenia.2–4 ,19 Cumulatively, 63% of respondents (17/27) reported that local practice did not fully correspond with current evidence-based guidelines. It is unclear from our research whether this was a conscious deviation from suggested practice. The current IDSA guidelines recommend restricting empiric use of glycopeptides to certain well-defined clinical situations where substantial benefit from such treatment is likely to be achieved. Appropriate indications for use of a glycopeptide in febrile neutropenia as defined by these guidelines are
pneumonia documented radiographically
positive blood culture for Gram-positive bacteria before full identification and sensitivity testing are available
clinically suspected serious catheter-related infection
skin or soft-tissue infection at any site
colonisation with methicillin-resistant Staphylococcus aureus (MRSA) or penicillin-resistant Streptococcus pneumonia
severe mucositis, if fluoroquinolone prophylaxis has been given and ceftazidime is used empirically.3
Persistent fever alone in a clinically stable patient is no longer considered to be an indication for the addition of a glycopeptide,3 but this was the most common practice among survey respondents. In addition, respondents listed central venous catheterisation in the absence of infective signs more often than haemodynamic instability or MRSA colonisation for prompting their decision to use teicoplanin upfront (figure 2). Again, this was in marked contrast to international guidelines. It seems that haematology units throughout the UK and Ireland have been reluctant, or at least slow, to apply the latest recommendations for the management of febrile neutropenia to local practice. NICE Guidelines were published in September 20124 and closely resemble IDSA Guidelines for targeted use of glycopeptides in febrile neutropenia.3 It is possible that greater local awareness of these guidelines may have a more pronounced impact on practice in the UK and Ireland than other publications prior to this point.
An examination of dosing and monitoring practice from our survey suggests that the manufacturer's recommendation as set out in the SmPC is heavily relied upon in UK and Irish hospitals, rather than TDM, as an indicator of therapeutic dosing in patients with haematological malignancy. Over 70% of respondents follow the manufacturer's SmPC recommended dose (18/25, 72% in febrile neutropenia; 34/44, 77% in documented infection) and only 12% (5/42) routinely monitor serum concentrations of teicoplanin in these patients. This is despite an abundance of evidence suggesting that conventional dosing as per the manufacturer's SmPC may be inadequate to produce therapeutically effective serum levels.10 ,11 ,15 The fact that all respondents whose practice differed from the SmPC recommendation used higher doses indicates that they agree with this evidence.
The relative lack of TDM when using this drug is of particular interest. TDM is recommended during teicoplanin therapy primarily to ensure therapeutically effective plasma concentrations are achieved and not merely to avoid toxicity.15 Using logistic regression analysis, Harding et al20 showed that the trough concentration of teicoplanin was a major predictor of treatment success in patients with S aureus septicaemia. Klastersky21 demonstrated that as the absolute neutrophil count decreases, higher bactericidal activity in the serum is required for successful treatment. Experimental animal data have shown that teicoplanin doses required to protect mice challenged with Staphylococcus haemolyticus were four times higher in immunocompromised than in normal animals.22 Furthermore, suboptimal concentrations of teicoplanin, especially early in the treatment period, can lead to the emergence of resistant Gram-positive organisms particularly when the minimum inhibitory concentration is close to the breakpoint, and clinical failure can occur.23 Given the link between serum concentrations and treatment outcome, it might be prudent to conduct TDM routinely when using teicoplanin in patients with haematological malignancy in order to avoid subtherapeutic levels.
Serum trough concentrations of >20 mg/L have been associated with improved outcomes in the treatment of endocarditis, musculoskeletal infections and pneumonia.24–26 In a retrospective analysis of 42 clinical cases of Staphylococcal infection, including skin and soft tissue infections, bacteraemia, endocarditis and bone and joint infections, patients with trough concentrations of ≤20 mg/L were shown to be less likely to be cured than patients with trough concentrations >20 mg/L.27
Higher doses than those currently recommended in the SmPC may need to be employed to achieve trough concentrations above a target of 20 mg/L.27 In a study of 141 clinically stable adults with bone and joint infection, Matthews et al28 found that increasing the daily dose from 400 to 600 mg statistically improves the chances of attaining trough levels >20 mg/L.
The case for higher doses and routine TDM is further strengthened when one also considers the altered pharmacokinetic behaviour of teicoplanin observed in patients with haematological malignancy. An understanding of these pharmacokinetic changes and how they may impact on drug concentrations is essential for the provision of dosing guidance. In particular, increases in the volume of distribution and clearance of teicoplanin have been observed in these patients and may result in lower than expected serum concentrations.10 In a population pharmacokinetic study, Lortholary et al10 estimated that 62% of patients with haematological malignancy receiving standard dosages of teicoplanin had low trough concentrations at 48 h. These findings were consistent with subsequent studies that also suggested that the manufacturer's SmPC dosing recommendations for teicoplanin may be too low for neutropenic patients with haematological malignancy.11 ,29
Increased renal clearance of several antibacterials has been observed in patients with haematological malignancy and may result in subtherapeutic plasma concentrations. This is particularly noted with renally cleared, hydrophilic antibacterials, like glycopeptides and aminoglycosides.10 ,11 Udy et al14 refer to this phenomenon as ARC, which describes enhanced elimination of circulating solute by increased glomerular filtration. Teicoplanin is eliminated almost entirely by glomerular filtration.24 In patients without organ dysfunction, ARC is thought to result from the inherent haemodynamic response to the disease state, together with common clinical interventions, such as aggressive fluid resuscitation, which together promote increased delivery of solute to the kidneys, increased renal blood flow and increased glomerular filtration.30 Pea et al11 hypothesise that in patients with acute leukaemia, at least early in the postchemotherapy period, the enhanced renal clearance of hydrophilic antibacterials might also be due to an increased glomerular filtration rate counteracting the huge renal load of protein-derived catabolites derived from lysis of circulating cells.
Teicoplanin is highly bound to plasma albumin (90–95%), and binding is linear with increasing serum concentration up to 300 mg/L.24 Hypoalbuminaemia, and a consequent decrease in binding capacity, will result in significant increases in the unbound fraction of teicoplanin in the plasma enabling more rapid distribution of free drug into interstitial spaces and increased renal clearance.11 These effects would translate into lower than expected plasma concentrations of teicoplanin for a given dose provided there is no renal impairment. A strong argument can therefore be made for the use of higher doses of teicoplanin in patients with haematological malignancy to account for altered pharmacokinetics as well as immunosuppression.
Our survey may not be representative of teicoplanin practice across all UK and Irish institutions with a haematology unit. We estimate that there are approximately 20 institutions with a haematology unit in Ireland and approximately 220 in the UK.9 Our survey was distributed to 168 different institutions, but some of these may not have had a haematology unit. Moreover, there may have been membership overlap across institutions and we cannot exclude the possibility that there may have been instances of multiple responses from the same hospital. Thus, only an estimated response rate of approximately 30% of surveyed units can be made, assuming that one submission was made by each institution. Nevertheless, the number of responses is large enough to give an indication of overall current practices.
This study suggests that there are considerable differences in usage of teicoplanin in patients with haematological malignancy across the UK and Ireland. Moreover, in most respondents’ centres, practice did not reflect current international recommendations for clinical indication. When teicoplanin was initiated empirically, most centres did not change the regimen even when a particular infection was subsequently documented. Current evidence suggests that dosing according to the manufacturer's SmPC recommendations may not be optimal. Despite this, the majority of hospitals continue to follow standard dosing recommendations and do not conduct TDM. It is possible that a lack of consensus on optimal dosing and monitoring practice may be a contributory factor, and further research to determine the optimal dosing and monitoring approach with teicoplanin in this unique patient population is required. Notwithstanding current uncertainty around the optimal dosing approach, considerable improvement in translating available evidence around indications for usage into regular practice could be more readily achieved. Clinical pharmacists could aid this process through education and increased interaction with physicians regarding drug therapy decisions. Such an approach may help avoid unnecessary glycopeptide usage and thereby reduce associated drug expenditure and bacterial resistance.
What is already known on this subject
Current ‘best practice’ for teicoplanin use in febrile neutropenia promotes restricted application to only situations where substantial benefit is likely to be achieved.
Routine addition of teicoplanin in a clinically stable patient with persistent fever before documentation of a Gram-positive infection does not improve outcomes.
Evidence suggests that the manufacturer's standard doses of teicoplanin may be inadequate for patients with haematological malignancy and therapeutic drug monitoring is considered mandatory to optimise therapy rather than to avoid toxicity.
What this study adds
Study findings suggest that despite an abundance of evidence for restricted use of teicoplanin in febrile neutropenia this approach is still not widely practised by UK and Irish institutions.
Heavy reliance on the manufacturer's standard dosing recommendations for teicoplanin predominates, rather than therapeutic drug monitoring, as an indicator of therapeutic dosing in UK and Irish centres.
The authors acknowledge the assistance of the British Oncology Pharmacy Association. The authors also wish to thank Maria Tallon, Felicity McDonnell, Joan McGillycuddy, Mary Coyle and Dr Mark Coyne for piloting the questionnaire.
Contributors All authors contributed to the writing of this manuscript.
Funding This study was supported by a grant from The Meath Foundation (all authors). The Meath Foundation has had no involvement in the present study.
Competing interests None.
Ethics approval Ethical approval was granted by the Tallaght Hospital/St James's Hospital Joint Research Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.