Nosocomial infections (Proceedings)

Nosocomial or hospital-acquired infections (HAI) are defined as infections that occurred during hospitalization. HAI lead to increased morbidity, mortality and costs. Rates of HAI vary by country, region, and facility. Typical rates are 7-10% for entire hospitals, and ~20% in ICUs (Bearman '06). Far less is known about HAI rates in veterinary medicine. A 2003 abstract reported a rate of 16.9 HAI/100 dogs; two dogs died as a result of their infection (Shaw '03). Not all HAI are resistant to antimicrobials. However the level of antimicrobial resistance increases with duration of ICU hospitalization. Fecal E. coli exhibited 68.7% resistance to ampicillin by day 6 of ICU hospitalization versus 19.6% on day zero. Levels of resistance also rose to ampicillin/clavulanic acid and enrofloxacin (Ogeer-Gyles '04).

Nonetheless it is the multi-drug resistant organisms that are more problematic causes of HAI. Sharp increases in antimicrobial resistance have been observed. Overuse of antibiotics is the main culprit and there is tremendous evidence for this, both in clinical medicine and in food animal production. Contrary to initial thought processes, reductions in antibiotic usage have not lead to decreased levels of resistance.

E. coli resistance to fluoroquinolones has continued to rise sharply despite reductions in prescribing. Veterinary reports confirm this trend with susceptibility decreasing from nearly 100% to less than 70%. A group of gram negatives that raise further concern are the extended spectrum beta-lactamase (ESBL) producers. These enzymes confer resistance to many of the cephalosporins. There are several classes of ESBL enzymes; CTX-M is one class which is becoming more prevalent allows Enterobacteriaceae to hydrolyse cefotaxime.

Methicillin-resistant Staph Aureus (MRSA) has actually been around since 1961, the year after methicillin came into use. It should be stated that we do not prescribe or microbiology labs directly test for methicillin susceptibility. Oxacillin is the in vitro testing agent. Among HAI of Staph, the proportion of methicillin resistance has risen dramatically; of staph isolates, it more than doubled from 1987 to 97 up to 56% according to the CDC. Regardless, news and media outlets attention MRSA has made it kind of a "rock star" amongst the multi-drug resistant pathogens. Most MRSA clones are now resistant not only to all β-lactams and cephalosporins, but also fluoroquinolones. The 2004 National Nosocomial Infection Surveillance system reported continued increases in resistant organisms in the ICU.

The other organisms of concern are vancomycin resistant enterococci, imipenem resistant Pseudomonas, and third generation cephalosporin resistant Klebsiella, Enterobacter, and Psudomonas.

Increasing rates of antimicrobial resistance coupled with the reality that inappropriate therapy in the intensive care unit for the first 24 hours of hospitalization can result in a doubling of the mortality rate. The challenge clinically is to not only know the likely organisms, but anticipate what their antibiograms will be.

Hand washing is far and above the easiest, simplest, most cost-effective solution to decreasing HAI. Our hands are the serve as the major fomite for organism transfer from patient to patient. Current hand washing rates vary by hospital and amongst staff. Observational compliance rates range from 5-81%. Nurses are generally more compliant than doctors. A targeted hand hygiene program in Melbourne increased hand cleansing rates from 21 to 42%; concurrently MRSA bacteremia decreased 57% (Johnson '05). Numerous reports echo similar results: hand washing significantly reduced HAI. Improving compliance is the challenge. Washing one's hands for two minutes with soap is time consuming and sinks may not be readily available. Several alcohol based hand hygiene solutions are available; some include chlorhexidine. These solutions can be available at convenient locations throughout a hospital, and require only a 30 hand rub. Gross hand contamination (i.e. obvious debris) should be removed by hand washing as the alcohol rinses will not penetrate adherent debris.

Instituting a surveillance program is a key step. The Netherlands have a very aggressive search & destroy policy for organisms such as MRSA. They routinely screen healthcare workers with culture; when a healthcare worker is identified as a carrier treatment which consists of chlorhexidine baths and mupirocin nasal rinses. Hence the level of MRSA HAI in the Netherlands is in the single digit percentiles. Our Clinical Microbiology Lab at the University of WI Teaching Hospital prints a report yearly which has in vitro susceptibility patterns for infections identified in the large and small animal hospitals. Programs can be costly and time consuming, but they do not have to be and can be made very practical. By retaining an additional copy of culture results from HAI infections in a folder, one can quickly build a data base of internal information. The information can be entered into an electronic data base and reviewed periodically. Of course appropriate sample collection and submission is the first step to identify any HAI.

Specific HAIs

Incisional infections

Surgical site infection (SSI) rates have been reported of 3.0-5.9% in veterinary medicine (Eugster S, 2004; Nichols M, 2002). Numerous risk factors are associated with development of a SSI including increased surgical or anesthetic time, endocrinopathy, intact male, increased personnel in the operating room, dirty surgical site, and use of propofol (Heldmann E, 1999). Appropriate perioperative antimicrobials have inconsistently been shown to reduce surgical site infections (Eugster S, 2004; Nichols M, 2002). In humans, risk factors for postoperative wound infection include duration of hospitalization, age, surgical technique, obesity, nutritional status, concurrent diabetes, appropriate antibiotic administration, surgery involving the abdomen, and duration of surgery.

Intravenous catheter related infections

Intravenous Catheter (IVC) related infections can develop by way of multiple mechanisms. Possibilities include migration of skin flora down the exterior catheter wall, contamination of the catheter hub and migration down the lumen, contamination of the infusate such as TPN, dextrose or blood products, and lastly hematogenous seeding of the IVC from distant sites. Migration of organisms down the exterior or lumen of a catheter occurs most commonly. The CDC has strict definitions of a catheter related blood stream infection (BSI): recognized pathogen cultured from the blood which is unrelated to distant site infection. Concurrent semi-quantitative culture of the catheter tip is recommended. Growth of similar organisms and sensitivity profiles from both the distant blood culture and the catheter tip confirm a catheter related BSI.

Very little information is available regarding incidence, morbidity and mortality related to IV catheter related infections in dogs or cats. One recent report (Marsh-Ng '07) identified 24.5% (37/151) positive catheter tip cultures. Thirty of thirty seven cultures were from dogs; 12 were from cephalic catheters; 19 from jugular, and 6 from saphenous IVC. Enterobacter was the most frequently cultured organisms (46%). Phlebitis at the catheter site was not associated with culture results. An abstract describing 25 cases of bacteremia identified an overall mortality rate of 32%, and of the gram negative infections, 62% were considered HAI (Abelson '05).

A group at the Ontario Veterinary College evaluated contamination rates of multi-dose vials (MDV) within the teaching hospital. They identified an 18% bacterial contamination rate of their MDVs. In a second phase of their report swabbing MDVs with 70% alcohol reduced the contamination rate to zero.

In people conservative estimates report deaths from blood stream infections as the 8th leading cause of death in the United States (Wenzel '01). Associated cost estimates are $45,000 per infection and $2.3 billion annually. Median rates of catheter related BSI are 1.8 – 5.2 per 1000 catheter days (Pronovost '06). Applying these median rates to our profession might seem impractical. However, a 20 cage ICU with average capacity of 50% with 50% of the cases having a CVC (jugular or saphenous) would result in 1,825 catheter days/year. Assuming similar rates would estimate 3.3 to 10 blood stream infections/year; in reality our rates of BSI may be even higher. Catheter related BSI rates can easily be prevented by simple and affordable procedures. The procedures include: appropriate hand hygiene; use of chlorhexidine for skin preparation; full barrier precautions (cap, mask, large drape, gown, sterile gloves); jugular vein is preferred; saphenous or femoral vein avoided when possible; and removal of any unnecessary CVC. These practices are easily adopted and should be followed.

Urinary catheter related infections

When compared with CVC BSI and ventilator associated pneumonia, urinary catheter related infections are much more tolerable. Rates reported in people are an average of 5% per day with nearly all being bacteriuric after a month. Rates of bacteremia originating from bacteruria are around 4%. Catheter related bacteriuria is usually asymptomatic, uncomplicated, and resolves following catheter removal. In other words, colonization is common and occurs at high rates; but urinary catheter colonization contributions towards morbidity and mortality are low. Outbreak investigations indicate that lack of proper hand hygiene is largely responsible for the transmission of these often multi-drug resistant organisms (Saint '99); urinary catheter related colonization and infection can serve as a reservoir of drug-resistant pathogens.

A couple reports are present documenting catheter related bacteruria. One group identified at 10.3% incidence rate; catheters were placed with sterile technique and continued hygiene (Smarick '04). A second group reported at rate of 19%, however the urinary catheters were in place longer (Ogeer-Gyles '06). Reports thus far have not reported if bacteruria was persistent following removal of the urinary catheter; nor has the urine sediment results been reported. It is important to differentiate between colonization and infection in patients with an indwelling urinary catheter.

Ventilator associated pneumonia

Unfortunately there are no published reports of ventilator-associated pneumonia (VAP) rates in veterinary medicine. Undoubtedly it occurs. The definition of VAP can be challenging in the critically ill patient who often has concurrent pulmonary pathology necessitating mechanical ventilation. Traditional criteria in people include fever, cough, purulent sputum, radiographic pulmonary infiltrates, and supportive culture or gram's stain. VAP doubles patient mortality rate, lengthens ICU and hospital stay, and increases hospitalization costs. Measures to reduce rates of VAP include hand hygiene and gloves, 30-45° backrest elevation, suctioning subglottic secretions, and oral chlorhexidine rinse. A consensus has yet to be reached for prophylaxis with histamine-2 receptor antagonists versus sucralfate, or for selective digestive decontamination.

Conclusion

Undoubtedly HAI occur in veterinary medicine. Precise rates have not been clearly described except for surgical site infections. Incisional infection rates can be reduced by decreasing use of propofol, and decreasing anesthesia and surgical times. Simple and cost effective procedures can easily be instituted to reduce rates of HAI including promotion or regular hand hygiene, and sterile technique with barriers when placing intravenous catheters. It is recommended to evaluate urine following removal of a urinary catheter. Regardless, it is imperative to differentiate between colonization and infection; clinicians should avoid prescribing antimicrobials for cases with bacterial colonization.

References available upon request.