Evaluation of the Q Score and Q234 Systems for Cost-Effective and Clinically Relevant Interpretation of Wound Cultures

The Gram stain has been used as a screening tool to assess the quality of various specimens, including respiratory, urine, and superficial wound specimens (4). The premises of this approach are that (i) polymorphonuclear cells are indicative of infection/inflammation and (ii) squamous epithelial cells are suggestive of superficial contamination. As a quality-screening tool, the Gram stain is an acceptable method for evaluating noninvasively collected lower-respiratory specimens for culture. The Gram stain may also be used to guide the extent of work-up of organisms in a mixed culture (1, 3). Wound cultures are often difficult to interpret due to the polymicrobial nature of these specimens, particularly when the specimens are improperly collected. The majority of wound specimens are collected on swabs, which can be easily contaminated with commensal flora from the surrounding mucosal and skin surfaces. Although most microbiologists agree on which organisms are considered pathogenic in wound cultures, there would likely be a difference of opinion on the extent of work-up, particularly if a culture contained more than two or three organisms. In an attempt to develop a more standardized and rational approach to the work-up of wound cultures, two protocols, the Q score system and the Q234 system, were compared to our current wound culture procedure (1, 2, 3). All three methods were also assessed to determine both their clinical relevance and cost-effectiveness.

The Q score was determined for each specimen by using the protocol outlined in Fig. 1. The Q score system applies positive values to the number of polymorphonuclear cells (PMNs) and negative values to the number of squamous epithelial cells (SECs) seen in a direct Gram-stained smear (1). These values are added, with the resulting figure representing the “Q score” which is used to assess the quality of the specimen and to determine the extent of culture work-up of potential pathogens (PP). It should be pointed out that this system bases the quality of the specimen upon the number of SECs present, not the absence of PMNs. The organisms considered PP for the Q score system included beta-hemolytic streptococci, Enterococcus sp., gram-negative rods, Bacillus anthracis, Staphylococcus aureus, and yeast (2). Up to three potential pathogens can receive identification (ID) and antimicrobial susceptibility testing (AST) from a good-quality specimen (Q3). The lower-quality specimens, Q1 and Q2, would have fewer PP fully worked-up (one and two, respectively). However, if the number of PP in the culture was greater than the Q score (e.g., Q2 with three PP), the results of the initial specimen's Gram stain were utilized. If the number of PP seen in the Gram stain was less than or equal to the Q score, ID and AST were performed for the PP seen in the Gram stain. If all PP were seen in the Gram stain, then morphological identification (MID) was performed. An example would be a Q2 specimen growing three PP in culture with one PP seen in the Gram stain. Only the latter PP would receive ID and AST. If all three PP were seen in the Gram stain, then MID of all PP would be done. Specimens with a Q score of zero (Q0) received limited morphological identification in this study, but these specimens could be discussed with the ordering physician as to their acceptability and possible recollection. MID consisted of the following rapid tests/observations: colony Gram stain, hemolysis on blood agar, lactose fermentation reaction on MacConkey agar, oxidase, catalase, staphylococcal latex (Slidex Staph kit; BioMerieux), streptococcal latex (Slidex Strepto-kit; BioMerieux), pyrrolidonyl arylamidase PYR (Hardy Diagnostics), and Murex Candida albicans (Remel).

The Q234 system is based on the number of PP present in the culture (2). The Gram stain is used as a guide to determine an organism's significance if more than two potential pathogens are present (Fig. 2). The organisms considered PP with the Q234 system were identical to those for the Q score system described above. If one or two PP were present in culture, ID and AST were performed on these organisms. If four PP were present in the culture, MID was performed on all four PP. If three PP were present in the culture, the results of the Gram stain were utilized; if one or two of the PP in culture were seen in the Gram stain, ID and AST were performed on these isolates; and if all three PP were seen in the Gram stain, MID was performed on all three. In addition, if moderately diverse floras with rare PP or very diverse floras with few PP were present, MID was performed on the PP since they were found in much reduced quantities compared to the commensal flora. Flora was defined as those organisms normally found on skin/mucosal surfaces, including diphtheroids, coagulase-negative staphylococci (CNS), nonpathogenic Neisseria species, and alpha- and nonhemolytic streptococci. The following definitions were used for growth quantities: rare, ≤10 colonies in primary area; few, >10 colonies in primary area; moderate, >10 colonies in secondary area but <10 colonies in tertiary area; and many, >10 colonies in tertiary area.

The current method of interpreting wound cultures at our university hospital (UH procedure), which was used during the study period, consisted of ID and AST if there were three or fewer organisms in the culture. These organisms included beta-hemolytic streptococci (no AST), Enterococcus sp., Bacillus anthracis, gram-negative rods, Staphylococcus aureus, coagulase-negative staphylococci, yeast, Erysipelothrix sp., and Corynebacterium jeikeium (catheter sites). Full biochemical ID was reserved for gram-negative rods, C. jeikeium, and Erysipelothrix. The remaining organisms received MID, with possible isolates of B. anthracis referred to a reference laboratory. No isolates of B. anthracis, C. jeikeium, or Erysipelothrix were seen during the study. About 50% of the coagulase-negative staphylococci which received AST were from pure cultures and most of the remaining isolates were found in culture with one of the additional PP listed above. If there were four or more organisms in the culture, the culture was reviewed with a supervisor and/or the director in order to determine further work-up.

A total of 305 wound cultures were evaluated by using the three approaches outlined above. The specimen sources are shown in Table 1. Normally sterile body fluids (e.g., pleural, peritoneal, and synovial) sent on swabs were not entered in the study. Specimens were collected with BD Culture Swab Plus and plated onto tryptic soy broth with 10% sheep blood, chocolate, MacConkey, and colistin-nalidixic acid agars. Plates were incubated at 35°C in CO₂ and held for a total of 72 h. Direct Gram stains were performed on all wound specimens. Smears were examined to determine the number of PMNs and SECs and the number and types of organisms present. Vitek2 was used as the primary identification and susceptibility system during this study with AST-GP61 and AST-GN04 cards used for AST. Reagent costs were determined for tests which would have been done or not done relative to those of the UH procedure. These costs pertained primarily to Vitek2 susceptibilities and/or identifications.

Of the 305 wound cultures evaluated, 147 (48%) and 84 (28%) would have been interpreted differently with the Q score and Q234 systems, respectively, relative to the interpretation with the UH wound procedure. The majority of these differences were the results of unnecessary culture work-up utilizing the UH procedure. In Table 2, a significant portion of the unnecessary tests included ASTs for gram-positive organisms, primarily staphylococci. There were also a number of gram-negative organisms that would not have had ID or AST performed using the Q score.

There were only 12 and 18 cultures, respectively, with the Q score and Q234 systems that would have had additional work-up performed relative to the UH procedure (Table 3). The majority of these cultures contained four or more organisms (mixtures of PP and non-PP) which received MID with only the UH procedure; however, the PP received ID and AST with both the Q score and the Q234 systems.

The reagent cost savings relative to the cost of the UH procedure were greatest with Q score, totaling $4,895 per year. The cost savings with Q234 were more modest at $1,503 per year. The cost savings would increase for both “Q” systems if technologist time were factored in as well.

Although the Q score system had a yearly savings of $3,392 over the Q234 system and would have eliminated more culture work-up, it had several issues that need to be addressed prior to implementation. If utilized, the Q score system would require physician approval for not culturing specimens with a Q score of zero (Q0). It would also require the training of technologists on all shifts to examine Gram stains and generate Q scores (the Q234 system did not require a different interpretation of direct smears). In addition, there were clinical issues that would need to be considered before implementation of this system: 65 cultures containing S. aureus (47 methicillin-susceptible S. aureus [MSSA] and 18 methicillin-resistant S. aureus [MRSA]) would not have had AST reported with the Q score system compared to the Q234 system and the UH procedure. In 26 (40%) of these cases, the Q score was Q0, indicating that the specimen would need consultation with the ordering physician prior to determining whether work-up was relevant or recollection was warranted. S. aureus was growing in pure culture (16 cultures with heavy growth) and was visible in the Gram stain of 17 (65%) of these 26 specimens. Not culturing these specimens or limiting culture work-up was thought to be unacceptable when these data were presented to infectious disease physicians at our institution. Only 11 S. aureus (9 MSSA and 2 MRSA) would not have had AST results reported by the Q234 system when such would have been reported as part of the UH procedure. All were from mixed cultures with at least three PP present.

Neither the Q score nor the Q234 system has indications for working up CNS. This resulted in the elimination of 49 CNS sensitivities with each system compared to the UH procedure. For 23 (47%) of these cases, CNS was growing in pure culture. In consultation with infectious disease physicians, it was thought that in cases where the specimen description indicated a significant site (sternal/chest, joint, or hardware), pure CNS should receive ID and AST. Our data for the 3-month period of the study revealed that 9 (39%) of the 23 cultures with pure CNS met these criteria.

The utilization of either the Q score or the Q234 system can help to minimize the likelihood of reporting misleading information to clinicians that might result in inappropriate antibiotic therapy as well as limit unnecessary work-up of mixed specimens that have little clinical relevance. One must keep in mind that exceptions to these protocols exist, and a physician may have a valid reason for deviation from these recommendations. The laboratory needs to remain open to discussion of the necessity to have further ID and/or AST performed on particular isolates.

With adoption of either the Q score or the Q234 system, allowances could be made for additional work-up when the specimen source or clinical history warranted. For example, additional work-up of gram-positive rods may be warranted to identify Corynebacterium jeikeium from catheter sites or Erysipelothrix. Consideration may be given to ID and AST of pure cultures of alpha-hemolytic streptococci from abscesses. Upon adoption of the Q234 system, we did include allowances for all of the previous scenarios. Finally, all isolates of S. aureus and enterococci could be screened for resistance to oxacillin and/or vancomycin if deemed necessary for clinical or infection control purposes. At this time, our institution does not require this type of comprehensive screening.

It should also be recognized that although the Q score system was originally developed for noninvasively collected lower-respiratory secretions, where the “rejection” of unacceptable specimens is now the standard of care, the rejection of wound specimens may not be acceptable without a consultation with the ordering physician. If using the Q score system for wound work-up, one could modify the approach to morphologically identify all PP in a Q0 specimen rather than labeling the specimen as not satisfactory for culture. This would alert the clinician to the PP that are present in the specimen (as does the Q234 protocol) and allow them to contact the laboratory if further work is warranted.

In conclusion, both the Q score and the Q234 systems showed clinical relevance and cost effectiveness and would allow for standardized approaches to the work-up of wound cultures. However, the Q234 system proved to be a more practical procedure to implement in our laboratory. The Q234 system does not require a change in Gram stain interpretation or physician approval for specimen rejection. Also, all PP are reported from a culture with either ID or MID, allowing for further consultation with the ordering physician if clinically warranted. Finally, the Q234 system is more structured and cost effective than our current method, was acceptable to our physicians, and will allow technologists to make more independent decisions about the significance of organisms in a wound culture.

• Open in new tab
• Download powerpoint

FIG. 1.

Determination of Q score.

• Open in new tab
• Download powerpoint

FIG. 2.

Q234 system flow sheet. *, If the potential pathogens are rare with at least moderately diverse normal floras, or are few with at least very diverse normal floras, the PP work-up is limited to MID with no AST. The following definitions were used for growth quantities:: rare, ≤10 colonies in primary area; few, >10 colonies in primary area; moderate, >10 colonies in secondary area but <10 colonies in tertiary area; and many, >10 colonies in tertiary area.

• Copyright © 2006 American Society for Microbiology