POLQM Quality Conference john.galbraith@viha.ca October 2 nd, 2017
Discuss the accuracy of results in microbiology, especially in susceptibility testing Review how well microbiology results are interpreted Consider opportunities for improvement
HIV 1+2 Ab + HIV p24ag (Screen): Reactive HIV 1+2 Ab + HIV p24ag (Supplemental): Reactive HIV 1 RNA (PCR/NAAT): Diagnostic HIV NAT is not performed on samples which are confirmed reactive by immunoblot. HIV 1+2 Ab (Immune Blot): HIV 1 Reactive HIV Ab Report: Findings indicate HIV 1 infection Is this report accurate? Is it clear?
HIV Ag/Ab Combination Test Result: Positive Comment: All positive serology samples are sent to the public health laboratory for confirmatory testing. The PPV of positive tests is greater than 99%. Island Health Communicable Disease Control will contact the MRP to assist with public health followup If you would like to request assistance, please contact.(1-866- )
62 yo male admitted with cord compression and destructive lesion at T8-9 Seen by Neurosurgery Post op diagnosis: T8-9 compression fracture secondary to infection or neoplasm
Gram: 1+ polys 1+ Gram positive cocci in pairs Prelim report at 5 days: no growth Final Report: Micrococcus sp isolated from enrichment media only. Culture results may represent contaminant or true pathogen. If clinically relevant please contact microbiology for further work-up.
Bone sample: Gram: no polys no visible organisms. Culture: no growth at 10 days Tissue sample: Gram: 2+ polys 1+ Gram positive cocci in pairs. Culture: no growth at 10 days and NEGATIVE for bacterial DNA by direct 16s PCR (performed at BCCDC)
Grew Streptococcus salivarius
Were the reports provided accurate? Were they clinically relevant? Could we improve?
Generally very accurate Some challenges in susceptibility testing
42/1896 (2.2%) of labs reported an MRSA challenge isolate as susceptible to cloxacillin/oxacillin 16/653 (2.5%) of labs reported a VRE challenge isolate as susceptible to vancomycin 99/1536 (6.4%) of labs reported a CRE challenge isolate as susceptible to carbapenems M Noble personal communication
It is no longer acceptable not to perform susceptibility testing on strains of S.pneumoniae 62% of category A labs reported SXT 66% reported a FQ 90% reported erythromycin 74% reported penicillin results
It is critical that vancomycin susceptibility always be reported with Pen-R strains
Most labs correctly reported pen S (but 3 reported oxacillin S) Most labs correctly reported erythro R Almost all labs (except 3) reported levo S without any comment/caveat therefore ungraded Only 9/13 reference labs reported SXT R so also not graded
This strain of S pneumoniae demonstrates first step resistance to ciprofloxacin. Even though in-vitro testing for the respiratory quinolone moxifloxacin tested sensitive, cross resistance within the antibiotic class is likely and can develop quickly. Fluoroquinolones should not be used when the strain has this mutation.
Accompanying Gram stain (G082): 4+ neutrophils, 1+ epi, 4+ Gram positive cocci suggestive of S.pneumoniae ID and sens UNGRADED as only 10/15 reference labs reported S.pneumoniae Overall 68 labs reported S.pneumoniae and 22 labs reported no growth
The correct MIC for cefotaxime and ceftriaxone was 2ug/mL = I (for nonmeningeal isolates)
CMPT advises an explanatory comment Note: The interpretation of susceptibility results for penicillin, cefotaxime and ceftriaxone varies depending on the presence or absence of meningitis. In the setting of possible infection of the central nervous system, clinicians should use the interpretations for meningitis.
1 yo with empyema All labs isolated and reported S.pneumoniae Intended penicillin MIC was 4 ug/ml = I for parenteral treatment of non-meningitis
Intended MIC for cefotaxime and ceftriaxone was 2 ug/ml = I (non-meningeal) * 3 labs reported R with MIC of 1.5 or 2 ** 3 labs reported I with MIC of 1.0 and 2 labs reported R with MIC of 2.0
CMPT recommends against the routine reporting of fluoroquinolone results in children. If labs choose to report fluoroquinolone results, then they should include a cautionary note indicating that fluoroquinolones safety in children has not been established
Accuracy alone cannot ensure that the result will be acted on appropriately
67% of initial empiric antibiotic Rx was appropriate 50% of the changes in therapy in response to the microbiology reports were appropriate Campo & Mylotte. Am J Med Sci 1988
Review of culture reports in 1050 hospitalized patients Only 7% of culture results resulted in a change of therapy no consistent logical use of bacteriological culture results, including antimicrobial sensitivity tests Edwards et al Arch Intern Med 1973
Information overload Reports more allied to the laboratory than to the clinician Confusion between accuracy and clinical relevance Common laboratory terms cause confusion
1. If your patient is getting better continue the same treatment 2. If your patient is not getting better change your therapy 3. If you do not know what you are doing don t do anything (Ask a friend for help!) 4. Never let you patient be seen by a surgeon (unless you are an ID doc!)
There is a significant association between the appropriate use of microbiological tests and appropriate prescribing for genito-urinary infections among general practitioners Kelsey et al. Br Med J 1996
Requisition design Better to list clinical scenarios rather than listing individual tests
What is the clinical syndrome? Asymptomatic, Routine screen Dipstick?UTI Dipstick + culture?nephritis Microscopy Courtesy Dr Michael Chan
Outpatient requisition forms were modified to provide clinical conditions and the appropriate tests : Original Requisition Redesigned Requisition Urine dipstick Routine screen (dipstick) Urine microscopy Rule out UTI (dipstick and Urine culture culture) Nephritis workup (casts) New requisition form for inpatient had the following statement added: Urine will be kept in the lab for 6 hours. Please call ext 12345 if microscopy needs to be added.
Number of analyses 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 95% reduction in the number of manual urine microscopy Urine Dipstick Urine Culture Urine Microscopy Chen. Clin Bioc 2016
Original annual cost (2007) a Scenario 1. Add automated analyzer b Scenario 2. Add a FTE b New protocol (2009) a Total FTE salary $279,809 $279,809 $346,494 $149,905 Urine microscopy supplies $15,649 $75,723 $15,649 $923 Total annual cost $295,459 $355,532 $362,143 $150,828 Actual annual savings b $204,704- $211,315 a Actual annual cost b Estimated annual cost (cost of automated cell counters was estimated to be $1.50 per test). Over 47,000 redundant urine microscopy have been eliminated annually! Chen. Clin Bioc 2016
Clinical history: catheterized, cloudy urine Culture: E.coli > 10*6 x 100 CFU/L Susceptibilities: Sensitive: cefazolin, cipro, gentamicin, nitrofurantoin Resistant: trimethoprim/sulfamethoxazole
Culture: E.coli > 10*6 x 100 CFU/L Comment: The urine of patients with indwelling catheters frequently becomes colonized. Unless the catheter is changed or the patient is unwell treatment is not required. Imprudent antibiotic use can If there is evidence of sepsis please contact (MM/microbiology lab). Abx susceptibilities are available if required. Cunney and Smyth. Antimicrob Agents 2000
Reported urine cultures: no growth (31%), no significant growth (21%), mixed growth suggests contam (20%) BUT Suppressed all positive urine culture results (28%) and reported a comment indicating that most positive in-patient urine cultures represented asymptomatic bacteriuria and to phone lab if ID and sens required AMMI-CACMID Conference2016
Patients with + culture Prebundle Asymptomatic Bacteriuria (AB) Pts with AB who received antibiotic therapy 61 % 50% 70% 19% Postbundle 72% reduction of antibiotic therapy for AB 38% reduction in urine culture test requests Smyth et al AMMI-CACMID Conference 2016
To determine significance of result requires knowledge of the patient s clinical status The report should encourage communication with the laboratory/medical microbiologist
Consider: Recommendations for therapy at each stage of reporting Adding interpretive comments (in the language of the one receiving the result) Keep it clinically relevant
Hypertext On-line advice Spectrum/Bugs and Drugs Smart antibiogram
Encourage two way communication with laboratory Employ terms that are readily understood Communicate clinically relevant results (probable infection or probable contamination) State when further clinical assessment required
When to withhold reporting antimicrobial susceptibilities? When and how to report susceptibilities especially in controversial circumstances? (eg. ESBL, IBL, ampc, CPO etc)
In its simplest form, evidence-based clinical practice means that it is no longer acceptable to function out of habit, because it sounds reasonable or because it has always been done that way. We require data, data derived from robust, objective studies, to either support, refute, or refine clinical practice.
Unfortunately, however, objective, systematic outcomes studies in clinical microbiology are almost nonexistent.
Every microbiology sample is a consultation Making sure the report it is accurate is good Making sure the report is clinically relevant and clear is better Making sure the report optimizes clinical outcome is best