Edited by:
Reviewed by:
*Correspondence:
This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Patients treated with I&D and antibiotics did no better than those treated with only I&D, regardless of the antibiotic activity. In the GP surveys, 89% reported I&D, with or without antibiotics, to be their preferred management. Only 29.9% of GPs would routinely swab abscesses.
Little is known about Australian general practitioners’ (GPs) management protocols in practice for skin abscesses and how this might be evolving with the changing epidemiology of
After a 2011 review, clinical practice guidelines from the Infectious Diseases Society of America (IDSA) continued to recommend I&D; specifically, that effective treatment requires incision, thorough evacuation of pus, and probing the cavity to break up loculations (
We have calculated that in 2006, 3–5% of patients presenting with community-onset
With CA-MRSA increasing in prevalence in Australia and elsewhere, there may be an argument for all lesions to be swabbed for microbiological culture and sensitivity (MC&S), not only to tailor treatment (as in complicated infections) but also for MRSA surveillance (
Our study set out to describe the management by community GPs of staphylococcal skin abscesses so we can better understand how often I&D is performed, if and what types of antibiotics are used, and whether antibiotics affect patient recovery. We also wanted to understand how often and for what reason GPs send swabs for MC&S.
Community-Onset
Shared households (204) were followed up at 3-monthly intervals for up to 2 years (recruitment period 2008–2011). At each visit, index patients and household members had swabs obtained from nares and axillae for
Of the 291 index patients, 137 had MRSA infections and 154 had MSSA infections. The majority were skin and soft tissue infections (86.2%), and 66 of the index patients had skin abscesses (50 MRSA, 16 MSSA) and are included in the present analysis. We extracted data on the doctor-reported data management (when remembered) as well as patient-reported management and outcomes for these 66 skin abscess infections; including whether I&D was performed or if and what type of antibiotics were prescribed. We also established the resistance profile of the infecting strain, and the number of days off normal activities and timing of infection resolution.
To understand the clinical decisions behind treatment practices, we conducted a cross-sectional survey of community GPs on their management protocols for
A patient presents to your clinic with a boil in their right armpit. Over the course of 3 days, the area has become increasingly reddened, tender and the center is raised and now forms a pus-filled head. There are no signs of systemic infection but there is evidence of localized infection. The patient has no other health problems and is not on any medication.
With this history, how would you treat the patient? (Select 1 or more)
No treatment
Incision and drainage (I&D)
I&D and antibiotics
Antibiotics only
Other (please specify)
If this boil was weeping or lanced*, would you swab this boil for culture and if so why?
Followed by a series of multiple choice and open ended questions on treatment choices and motivators for, and frequency of, collecting swabs.
∗
We surveyed a random sample of 41 doctors treating community-onset
To determine how representative COSAHC doctors’ patient management was, we compared their survey responses to those of a random sample of 39 GPs practicing within metropolitan Melbourne who had ordered routine blood tests (Full blood count and Urea and Electrolytes) in the same period from the same pathology provider.
Both groups of doctors were invited to return a survey tool by fax or complete over the telephone. The exclusion criteria were being on leave for more than one month at the time approached and/or no longer working at the practice. A sample size of 40 per group was recruited to provide 80% power (α = 0.05) to detect a 30% difference in antibiotic use between GPs.
The survey tool was developed to assess management protocols and swabbing practice in the context of a short case study of a patient with an axillary skin abscess. We also asked participants to provide information on any change to their practice over the previous three years.
Descriptive statistics were applied, including computing differences between proportions and the associated
Abscess management and outcome information was obtained from the 66 index patients at the first COSAHC household visit. The susceptibilities of the organisms were known as the patients were recruited on the basis of a positive
Infection sites.
Site of infection | MSSA ( |
MRSA ( |
---|---|---|
Leg/foot | 31% (11, 59) | 34% (21, 49) |
Torso (front/back) | 19% (4, 46) | 24% (13, 38) |
Arm/hand | 19% (4, 46) | 12% (5, 24) |
Axilla | 13% (2, 38) | 12% (5, 24) |
Head/neck | 13% (2, 38) | 8% (2, 19) |
Groin | 6% (0.2, 30) | 10% (3, 22) |
Abscess management by the GPs, as reported by COSAHC study patients, are shown in
Patient-reported GP abscess management by methicillin resistance (causal organism).
Treatment | MSSA ( |
MRSA ( |
Total ( |
---|---|---|---|
I&D and antibiotics | 38% (15, 65) | 36% (25, 53) | 36% (25, 49) |
I&D only | 0% | 0% | 0% |
Antibiotics only | 25% (7, 52) | 30% (18, 45) | 29% (18, 41) |
I&D, antibiotics and other∗ | 25% (7, 52) | 24% (13, 38) | 24% (15, 36) |
Antibiotics and other∗ | 13% (2, 38) | 10% (3, 22) | 11% (4, 21) |
Patient-reported infection resolution is shown in
Patient-reported infection resolution by first home visit (by management protocol).
MSSA ( |
MRSA ( |
Total ( | ||||
---|---|---|---|---|---|---|
Treatment | Resolved | 95% CI | Resolved | 95% CI | Resolved | 95% CI |
I&D and antibiotics | 5/6 (83%) | (36, 100) | 15/18 (83%) | (59, 96) | 20/24 (83%) | (63, 95) |
I&D only | 0 (0%) | – | 0 (0%) | – | 0 (0%) | – |
Antibiotics only | 3/4 (75%) | (19, 99) | 12/15 (80%) | (52, 96) | 15/19 (79%) | (54, 94) |
I&D, antibiotics and other∗ | 3/4 (75%) | (19, 99) | 11/12 (92%) | (62, 100) | 14/16 (88%) | (62, 98) |
Antibiotics and other∗ | 2/2 (100%) | (13, 99) | 2/5 (40%) | (5, 85) | 4/7 (57%) | (18, 90) |
The antibiotics prescribed were not known for all patients, meaning the patient could recall the actual name of the antibiotic(s). However, the majority (85%) could remember the name of at least one of their prescribed antibiotics: 43 patients with MRSA and 13 patients with MSSA. We analyzed the antibiotic susceptibility profiles of the clinical isolates and the activity of the known antibiotics that were prescribed for these 56 patients (
In total, over the course of their infections, there were 18 prescriptions of known antibiotics given to 13 patients with MSSA, of which three (16.7%) were inactive. Of the 81 known antibiotics prescribed for 43 patients with MRSA, 50 (62%) were inactive. Overall, approximately 53.5% of prescribed antibiotics were inactive: these included penicillin/amoxicillin for both MSSA and MRSA, and cephalexin, flucloxacillin, augmentin, or dicloxacillin for MRSA infections.
Prescriptions for MSSA infections were dominated by penicillin 16.7% and flucloxacillin, dicloxacillin, cephalexin, and augmentin (66.7%); However, augmentin and penicillin are not recommended first line of treatment for MSSA infections: augmentin because of its unnecessarily broad spectrum activity and penicillin/amoxicillin because only 5–10% of MSSA are susceptible (
Of the 43 patients with MRSA infections who recalled at least one of their specific antibiotic(s), 35 (81%) were initially prescribed an inactive drug. Of these, 23% never received an active antibiotic, whilst the rest were subsequently prescribed active (58%) or unknown (19%) antibiotic(s) at follow-up visits. Three (23.1%) of 13 patients with MSSA who recalled the specific antibiotic(s) were also prescribed an inactive drug, with one subsequently prescribed an active antibiotic.
In our study, 56 patients were able to remember a total of 99 antibiotics (an additional 19 were remembered as a prescription but not by name). Thirty-seven patients (6 with MSSA and 31 with MRSA infections) were prescribed more than one antibiotic and, of those, thirteen (1 MSSA patients and 12 MRSA) were prescribed more than two antibiotics. In addition, eight were prescribed Mupirocin (site of use unspecified but all organisms susceptible). We presume that for every new antibiotic prescribed the patients required a repeat visit to their GP for assessment, although we do not know if the subsequent prescription changes were due to treatment failure or antibiotic change based on lab results showing resistance.
One hundred and seventy GPs were approached and 81 completed surveys (48%; 41 COSAHC GPs, 39 comparison GP and one unlabeled form that could not be ascribed to a particular GP group). The response rate was higher in the COSAHC GP group (60% compared with 38%). The average number of years in practice of those participating was 23.1 years. No differences were found with treatment preferences between the two cohorts (
Self-reported management practices by doctor group.
Treatment | COSAHC GPs 41 | Non-COSAHC | Total GPs (81) |
---|---|---|---|
GPs (39) |
|||
I&D and antibiotics | 29 (71%) | 27 (69%) | 57 (70%) |
I&D | 8 (20%) | 7 (18%) | 15 (19%) |
Antibiotics only | 3 (7%) | 5 (13%) | 8 (10%) |
Other | 1 (2%) | 0 (0%) | 1 (1%) |
Overall, GPs report ‘I&D and antibiotics’ as their preferred management of abscesses (70%), followed by ‘I&D only’ (19%) and ‘Antibiotics only’ (10%). Some specified I&D included additions; one would incise with review and another would prescribe antibiotic cream alongside. Other selections included ‘no initial treatment, but subsequent patient review’ (indicated as ‘other’ in
When we stratified treatment choices according to the GPs decision to swab, we found abscess management did not differ significantly between the two surveyed doctor groups, with 70% of both groups reporting their preferred management protocols as ‘I&D and antibiotics.’ This indicates we have not introduced patient-management bias via the recruitment method used, other than including a greater proportion of patients treated by GPs who swab. More COSAHC doctors did indicate they would swab if the abscess was weeping or incised (76.9% vs. 56.4%;
Overall, a minority (29.9%) of GPs reported that they would routinely swab abscesses for culture. Doctors reported they would swab for definitive diagnostic purposes (68.8%) or for persistent abscesses (97.4%). About half of doctors (48.6%) report being influenced by abscess size and a perceived increased risk of systemic infection or spread to adjacent tissues. When abscess size was identified as important, thresholds of size varied from >1 cm to >3 cm. Swabbing practice was not related to treatment choice or the commencement of antibiotics.
The rise of MRSA in the community was identified as reason for changing practice (23%). Some doctors report they are now more likely to wait for results to prescribe antibiotics and choose to swab based on clinical site and severity of abscess.
Eight doctors of COSAHC patients with abscesses also completed the GP survey. Whilst the numbers are small, it does provide us the added opportunity of direct comparison between a patient’ reported abscess treatment and their doctor’s survey responses (
Community GP-reported management for abscesses versus patient-reported management.
GP survey response | Patients’ report of treatment by same GP |
---|---|
Antibiotics | Antibiotics |
Antibiotics | Antibiotics |
I&D and Antibiotics | I&D and Antibiotics |
I&D and Antibiotics | Self I&D, Antibiotics |
I&D and Antibiotics | I&D attempt, penicillin injection, oral antibiotics |
I&D and Antibiotics | I&D ×3, antibiotic injection, oral antibiotics |
I&D and Antibiotics | Antibiotics (I&D of previous boil) |
I&D and Antibiotics | Antibiotics |
None of the eight responded with the recommended first line treatment for an abscess, regardless of size. Two reported that they would not perform I&D and would solely prescribe antibiotics, which corresponds to the patients’ report. The remaining six doctors who selected ‘I&D and Antibiotics’ in the survey varied in their actual clinical response – three out of six (50%) did not perform I&D and only prescribed antibiotics. Interestingly, two out of six (33%) administered an antibiotic injection as well as oral antibiotics. This is particularly surprising because there are no intramuscularly injectable antibiotics that are active against most
We live in an era of rising antibiotic resistance. Conservation of antibiotics is of critical importance, and in response antibiotic stewardship programs are being progressively implemented. Antibiotic stewardship originated to address the problems of antibiotic overuse in hospitals, but is now being extended to the community as it is increasingly realized most antibiotics for human use are prescribed and consumed in the community.
Australian GP compliance with clinical guidelines for abscess treatment has not previously been researched. In the US the management of community-acquired abscesses have been investigated both in general practice and in hospital emergency departments (
Similarly, in the SNOCAP-USA and DARTnet study (
STARnet (
Hospital emergency department research in the US showed only 17–19% of abscesses were treated with I&D alone and treatment with antibiotics alone ranged from 4 to 17%, while a combination of I&D and antibiotics was the most common practice (66–79.9%;
Our study focuses on community-onset abscess infections where MRSA is over-represented as a deliberate part of our sampling strategy. The data drawn from the prospective cohort of community-onset staphylococcal infections (COSAHC) identified via a private community-based pathology service, and the cross-sectional survey of doctors ordering tests through the same pathology service, has provided the opportunity to explore doctors’ attitudes and practices in the context of the changing epidemiology of
The antibiotic susceptibility of isolates for all the COSAHC participants were known and as this was the criteria for entry into the study. In the COSAHC study, we found that 85% of community-onset
Furthermore, where the antibiotic was known, the majority of patients with abscesses caused by MRSA (81%) were initially prescribed inactive antibiotics, compared with 23% of those with MSSA. Antibiotics were changed for 37 patients, resulting in nearly all MSSA (85%), but only 60.5% of MRSA infections ever being prescribed an active antibiotic. Overall, 53.5% of antibiotic choices were inactive against the targeted pathogen.
We have previously estimated that about 8% of infections caused by
From our survey we found that only 30% of community GPs swab for MC&S. Therefore, many doctors would be unaware of the proportion of antibiotics prescribed that were ineffective and they are unlikely to modify their prescribing practice. According to both contemporary and Australian guidelines, almost all of the prescriptions for uncomplicated abscesses could be considered unnecessary. In support of these guidelines, our study showed there was little difference in infection resolution whether the antibiotics were active or not. This was true for ‘I&D + antibiotics’ and ‘antibiotics’ alone (
If an antibiotic is prescribed it is usually at first presentation. The swab will not help in the initial antibiotic prescription but a MC&S may be useful in the case of treatment failure and MRSA surveillance. If I&D is performed correctly, our findings support guidelines and no antibiotics are needed and can be stopped. Patients’ expectations regarding treatment is a challenge for GPs. Patient understanding of best practice can be narrow and limited to their own situation. There is a need to increase the health literacy of patients and to make them aware of the population wider risks associated with antibiotic over prescription.
Although I&D was consistently reported by GPs to be their preferred treatment option for uncomplicated abscesses, the COSAHC study observed only 60% of patients with abscesses were treated with I&D in practice. Furthermore, I&D was always accompanied by antibiotics. This suggests that whilst there may be an understanding of the importance of I&D, there may be a lack of confidence in performing the procedure without antibiotic cover.
Given the number of antibiotic changes (37 patients requiring additional antibiotics) and therefore the presumed increased number of visits to the GP, it is clear that skin abscesses are difficult to treat and many different antibiotics are used.
Our study has a number of limitations. The response rate of 48% in the GP survey has the potential for sampling bias. However, we believe that any selection bias that may have been introduced would act to favor doctors who are more aware of community MRSA and therefore more keen to participate. Therefore, it is possible that these results underestimate the actual deviation from recommended practice in the wider GP community. If this were the case, then the true situation regarding GP practices may be even more removed from therapeutic practice guidelines than captured here. It would be useful to conduct a larger study to more comprehensively examine GP practice, and gain more detailed insight into GPs’ ability to perform I&D correctly. Incision type, use of pain control, irrigation, wound cultures, and packing, would inform targeted strategies (
Our findings demonstrate many GPs are not following guideline recommended practice when it comes to treating patients with staphylococcal skin abscesses. Our findings support the recommendations in the current antibiotic guidelines that would act to curb this trend (
Antibiotics not indicated unless spreading cellulitis with systemic symptoms.
Perform I&D correctly on all abscesses, even if antibiotic therapy is considered.
Modify therapy based on clinical response to initial therapy and the results of cultures and susceptibility testing. If
Overall, the doctors we surveyed demonstrated awareness of the changing epidemiology of
CP initiated this work, lead the study design and ethics approval, and managed the recruitment and oversaw the data analysis, and instigated the writing of this manuscript. GW was responsible for managing the partnership with Dorevitch laboratory and GP recruitment and contributed to the design of the survey tool, interpretation of data, and manuscript preparation. EB assisted in data analysis and preparing the figures for this paper. GC was responsible for the
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.