Prevention of Infection in Patients with Cancer

174 Seminars in Oncology Nursing, Vol 23, No 3 (August), 2007: pp 174–183

OBJECTIVES: To provide clinicians with the most reliable, updated evidence to support clinical decision-making PREVENTION OF and improve outcomes for patients with cancer who are at increased risk for infection. DATA SOURCES: INFECTION IN Review of two evidence-based sum- maries of prevention of infection interventions published by the On- PATIENTS WITH cology Nursing Society; MEDLINE and guidelines.gov literature review. CANCER CONCLUSION: Handwashing is the most important intervention to prevent infection in patients with cancer. Guide- lines-based intravascular catheter CHRISTOPHER R. FRIESE care and preventive activities can reduce infection incidence in this vulnerable patient population. Un- NFECTION refers to the symptoms caused by the multipli- derstanding risk factors for aggressive pathogens can help identify cation of microscopic organisms and subsequent invasion of 1,2 patients for rapid surveillance and I the body’s natural barriers. These organisms may be of isolation procedures. Additional bacterial, fungal, viral, or parasitic origin. Patients with cancer, multi-site research is required in compared with other clinical conditions, are at increased risk for oncology settings to recommend infectious complications. The factors that predispose patients with recent interventions for practice. cancer to infection include decreased supply and/or function of IMPLICATIONS FOR NURSING lymphocytes or granulocytes, wounds following invasive surgery, PRACTICE: the placement of vascular catheters, nutritional deﬁciencies, and 3 Oncology nurses should assess pre-existing or newly acquired comorbidities. In many cases, their adherence to evidence-based patients present with many or all of these factors at once. Preven- guidelines on infection prevention. tion of infection occurs through activities by patients, nurses, Outcomes are optimized when cli- physicians, and public health professionals to reduce the likeli- nicians identify high-risk patients hood of microbial multiplication and invasion.1 Despite the recog- and provide scientiﬁcally supported nition that patients with cancer are at increased risk for infection, interventions. preventive interventions often vary by setting and lack an evi- KEYWORDS: dence base. Infection, outcomes, cancer, The white paper on nursing-sensitive patient outcomes pub- interventions lished by the Oncology Nursing Society (ONS) recognized prevention of infection as an important safety outcome that is sensitive to nursing interventions.4 Four of the 15 outcomes endorsed by the National Quality Forum as nursing-sensitive performance mea- Christopher R. Friese, RN, PhD, AOCN®: sures are related to infection.5 Reimbursement strategies for Center for Outcomes & Policy Research, Dana-Farber Cancer Institute; Cancer Pre- health care services based on quality measures, termed “pay for vention Fellow, Harvard School of Public performance,” are anticipated to increase widely. Moreover, on- Health, Boston, MA. Address correspondence to Christopher R. cology nurses have an interest in preventing infection in patients Friese, RN, PhD, AOCN®, Harvard School of with cancer to facilitate timely anti-cancer therapies, promote Public Health, 44 Binney St. SM 271, Boston, quality of life, and improve patient satisfaction. It is with these MA 02115. Fax: (617) 344-0427. motivations that this article discusses the recent evidence base for interventions to prevent infection. The goal of this review is to © 2007 Elsevier Inc. All rights reserved. provide clinicians with the most reliable, updated evidence to 0749-2081/07/2303-$30.00/0 support clinical decision-making and improve outcomes for pa- doi:10.1016/j.soncn.2007.05.002 tients with cancer who are at increased risk for infection. PREVENTION OF INFECTION IN CANCER PATIENTS 175

Two principal documents were used to frame from phlebotomy. If the culture from the catheter this review. The first was a review completed by source is positive first, the difference in time to the author in 20041 as part of the initial ONS effort positive result is compared between the catheter to measure nursing-sensitive patient outcomes. and the phlebotomy sample. When the difference The second document is a summary of the 2005 exceeds 120 minutes, catheter-related infection is ONS Putting Evidence Into Practice (PEP) project more certain. Clinicians continue to debate the team to categorize interventions based on the frequency, number, and source(s) of blood cul- quality of the available evidence.6,7 Finally, com- tures required in the febrile patient with cancer. mon clinical problems, such as infections related Although no published studies answer the ques- to vascular catheters, pneumonia, and aggressive tion empirically, a sound protocol was developed organisms (Clostridium difficile,vancomycin-re- by Penwarden and Montgomery.9 The steps in- sistant Enterococcus [VRE], and methicillin-resis- clude: (1) obtaining the cultures within 30 min- tant Staphylococcus aureus [MRSA]) were ad- utes of fever; (2) obtaining one peripheral set of dressed specifically because of their high cultures with the initial fever, with cultures ob- prevalence and potential for poor outcomes in the tained only via the vascular catheter during sub- oncology patient population. This review con- sequent febrile episodes; and (3) cultures ob- cludes with knowledge gaps and suggestions for tained from each lumen of the catheter, with no future research. blood discarded. While individual scenarios may dictate a different response by clinicians, a stan- EVIDENCE-BASED SUMMARY OF NURSING- dardized approach to blood cultures improves the validity and reliability of the diagnostic test. SENSITIVE OUTCOMES:PREVENTION OF Catheter-related blood stream infections have INFECTION been measured using standardized terminology endorsed by the Centers for Disease Control and he first review conducted for the ONS focused Prevention (CDC), among other groups.10 The Ton: (1) available measures, and (2) effective number of infections per 1,000 catheter days is a interventions with a high level of evidence for the widely used metric in the literature. For example, prevention of infection in patients with cancer.1 A assume in 1 month a clinic cared for 100 patients computerized search of the Cumulative Index to who had their catheters for 3 days each (300 Nursing & Allied Health Literature (CINAHL) and catheter days/month). In that month, 4 patients the United States’ National Library of Medicine acquired a blood stream infection. The catheter- bibliographic database (MEDLINE) was conducted related blood stream infection rate would be using the search terms “registered nurses” and 4/300 ϫ 1,000 ϭ 13.33. While the calculation may “infection” or “infection control.” Searches were not be intuitive it accounts for the fact that differ- limited to papers published between 2000 to 2003, ent clinics vary in their patient volumes and the and classified as systematic reviews or meta-anal- length of time catheters are used. yses. Meta-analyses are quantitative syntheses of The reviewed literature was then categorized separate but related research studies. In addition, into five domains of infection prevention interven- the Agency for Healthcare Research and Quality’s tions: (1) hygiene, (2) intravenous therapy, (3) online database of clinical practice guidelines nutrition/gastrointestinal, (4) environment, and (www.guidelines.gov) was searched using similar (5) chemoprevention. Readers are referred to the terms. source document for a complete discussion of the The findings showed scant attention to mea- review findings. However, important findings are surement techniques of infection. While this is presented in Table 1. partly because of the heterogeneity of infection, The 2004 review concluded that several inter- very little empirical research is available on how ventions had strong support to recommend wide- clinicians or researchers should measure infection spread adoption, but many common practices, from the quality of care perspective. One paper such as diet modifications and “protective isola- addressed the concept of time to positivity to tion” of patients with neutropenia, lacked strong infection as a tool to discern vascular catheter- evidence. Measurement of infection was varied related blood stream infection.8 In this approach, and few clear approaches were available for clini- paired blood cultures are obtained; one from the cal use. Specific gaps in the infection prevention distal port of the vascular catheter, and a second literature were noted in the outpatient oncology 176 C.R. FRIESE

TABLE 1. Major Findings from the 2004 Prevention of Infection Review1

Domain and Intervention Findings Discussion

Hygiene Handwashing The single most important nursing intervention Antimicrobial soap and water, or the use of to prevent infection alcohol-based hand gel have equal efficacy Oral care Frequent oral care, including gentle The frequency of oral care and the ideal toothbrushing and flossing (as tolerated) are oral cleansing agent are unknown effective Intravenous therapy Catheter placement Avoid placement of catheters when patient is Not always possible in certain clinical functionally or quantitatively neutropenic circumstances. Aseptic practice and full barrier precautions during placement are recommended Injection port cleansing Injection ports on intravenous tubing or Sufficient time between swabbing and vascular catheters should be cleansed with access should be maintained to assure 70% alcohol before access the alcohol has dried Nutrition/gastrointestinal Nutritional route Enteral nutrition is preferred to the parental Increased risk of blood stream infections route has been observed for patients receiving parenteral nutrition Diet modifications There is little evidence for the “neutropenic Historical anecdotes have led to restriction diet” of fresh fruits, vegetables, and other food sources. Two randomized trials are currently investigating this question Environment Protective isolation No rigorous data exist to support protective These data differ from the CDC-endorsed isolation practices for patients with cancer approach of isolating patients with active and neutropenia infections to protect other immunocompromised patients Chemoprevention Prophylactic therapy Patients receiving induction therapy for acute Several agents are available to achieve for patients with leukemia should receive routine prophylaxis anti-fungal and anti-viral prophylaxis acute leukemia for fungus and herpes simplex. Patients with acute lymphocytic leukemia should also receive Pneumocystis carinii prophylaxis

setting, and for infections outside the blood- the prevention of infection project team.1,6,7,11 stream. These findings were shared with advanced Briefly summarized, a panel of staff nurses, APNs, practice nurses (APNs) at the ONS-sponsored APN and a doctorally prepared researcher conducted a retreat in 2004. While reviewers found the docu- literature search of a broader scope than the 2004 ment helpful, they requested additional detail for review. Reviewed manuscripts were not restricted interventions commonly used in practice with to systematic reviews or meta-analyses; lower lev- varying levels of evidence. These comments stim- els of evidence were included. To discriminate the ulated the ONS’s PEP initiative. evidentiary support for particular interventions, each manuscript was assigned a level of evidence THE PUTTING EVIDENCE INTO PRACTICE (highest reflecting randomized controlled clinical (PEP) PROCESS trials with 100 or more patients; lowest evidence level referring to clinical observation or editori- series of manuscripts and documents are al).12 Papers describing the outcomes of a specific A now available to describe the process and intervention were then aggregated by the team to findings of the PEP initiative and specifically for assign an overall level of evidence for the inter- PREVENTION OF INFECTION IN CANCER PATIENTS 177

vention to prevent infection. As a practical exam- more recent standards have been made available. ple, the prevention of infection team reviewed five Oncology nurses are advised to consult with their separate studies of varying levels of evidence to subspecialty colleagues, the Infusion Nurses Soci- conclude that chlorhexidine was not recommended ety (INS), for recent updated evidence-based stan- to prevent mucositis in patients with cancer. dards on a variety of vascular access devices. In Although the 2005 PEP Prevention of Infection 2006, the INS published standards of practice on a document reviews a more comprehensive array of wide variety of relevant topics to oncology nurses, interventions, significant gaps remain. Controver- including catheter site selection, skin preparation, sies remain in oral care (frequency, choice of dressing application and changes, flushing, and agent), isolation practices, and the use of fluoro- stabilization procedures.14 Both the ONS guide- quinolone prophylaxis. The literature is lacking in lines and the INS standards are supported by evaluation of standardized patient and provider published sources, however, they lack a level of education on prevention of infection for patients evidence weighting assignment. Thus, random- with cancer. While frequently hypothesized, there ized trials, as well as observational studies, sup- is no evidence that detailed infection prevention port the guidelines and standards as currently educational content improves outcomes. This is a published. Selected INS standards are presented high priority area for future nursing research. in Table 2. However, the reader is strongly encour- aged to review the entire issue of the journal for RESEARCH FINDINGS ON SELECTED TOPICS complete details and considerations for special patient populations. Important teaching points from the standards include frequent observation he reviews described above provide a general of the junction between the access device and the approach to the prevention of infection in the T patient’s skin. Daily inspection and palpation is patient with cancer. Additional specific areas are required, and dressings should avoid covering the worthy of attention. Infections related to vascular junction to permit inspection. Clinical practice catheters and pneumonia are highly prevalent and varies in the schedule of catheter and dressing are the source of significant morbidity and mor- changes. Basic principles include reducing the tality in the oncology patient population. In addi- number of “breaks” to the intravenous system tion, the increase in incidence of aggressive organ- where infectious agents could be introduced inad- isms places current and future patients at risk for vertently. poor outcomes. These specific topics were re- In addition to the published standards, the INS’s viewed in MEDLINE and guidelines.gov for the journal, Journal of Infusion Nursing, has numer- years 2000 through 2006. Restrictions were ous nurse-led, data-based research articles docu- placed on manuscripts in English and focused on menting outcomes following changes related to adult patients. When 300 or more manuscripts infusion nursing interventions. These articles are were retrieved, the search was combined with an excellent model for oncology nurses to emulate neoplasm as a search term to restrict to the on- in making changes to their own practices, as it not cology patient population. only increases nursing scholarship, but informs the community about novel, effective interven- Vascular Catheters tions for the patient population. Standardized, evidence-based interventions for One recent nurse-led study in a large homecare the patient with vascular access are vital to infec- infusion service reported 7 years of data on 551 tion control efforts. When providers deliver these patients.15 Using the standardized metric in cath- interventions consistently, outcomes are easier to eter-related outcomes, 20,879 catheter days were track and measure. Many oncology nurses are observed. This prospective, observational study aware of the Access Device Guidelines published reported 0.77 infections per 1,000 catheter-at- by the ONS.13 This guideline summarizes perti- home days. In addition to the outcome data, fac- nent issues related to the insertion, care, and tors associated with the infected catheters were maintenance of vascular access devices. The reported; total parenteral nutrition was associated guidelines summarize available evidence sur- with 6 of the 16 documented infections in the rounding controversial issues, including dressing study period (␹2 28.14, P Ͻ .001). Other risk type, cleansing agents, flushing agents, occlusion, factors included neutropenia, immunosuppressive infection control, and catheter removal. However, agents, multi-lumen catheters, history of catheter 178 C.R. FRIESE

TABLE 2. Selected Standards of Practice from the Infusion Nurses Society14

Topic Standard

Placement 1. Maximal barrier precautions (sterile gown, sterile gloves, cap, mask, eyewear, sterile drapes/towels) used for central, midline, and peripherally inserted central catheters 2. Accepted antiseptic solutions include alcohol, chlorhexidine gluconate, povidone-iodine, and tincture of iodine. A combination of alcohol followed by chlorhexidine gluconate or povidone-iodine (after the alcohol dries) are preferred Catheter site care 1. The catheter dressing change procedure should include cleaning the junction of the catheter and skin with antiseptic agents (detailed above) using aseptic technique 2. Sterile gloves and mask should be used for catheters with extended dwelling times, are placed centrally, or in immunocompromised patients 3. Sterile dressings should be applied to vascular access devices. Options include gauze that prohibits direct observation of the entry site (change every 48 hours); gauze with a transparent semipermeable (TSM) dressing (change every 48 hours), or TSM dressing alone (at least every 7 days) 4. The junction of the skin and catheter should be inspected and palpated for tenderness daily 5. Catheters should be stabilized to restrict movement. Manufactured stabilization devices are preferred to sterile tape or surgical strips. If the latter are used, they should not be placed on the catheter-skin junction 6. Implantable ports: safety non-coring needles should be changed every 7 days. If a TSM dressing is used with gauze to anchor the needle and the catheter-skin junction is visible, dressings should be changed every 7 days. Needle and dressing changes should be timed simultaneously Intravenous administration 1. Sets should be changed every 72 to 96 hours, or whenever contamination or sets compromised integrity of the system is suspected 2. Aseptic technique should be used for set changes 3. Changes should coincide with new catheter placements, extension sets, ﬁlters, stopcocks, and/or initiation of new containers of solution 4. All sets should use luer-lock design 5. Sets used for parenteral nutrition should be changed every 72 hours or immediately after suspected contamination or compromised integrity 6. Sets for intravenous fat emulsions should be changed every 24 hours 7. Sets used for blood products should be changed after each unit or after four hours, whichever comes ﬁrst 8. Hemodynamic monitoring sets should be changed every 96 hours

infection, and serum albumin Յ 3.5. This study 3.4/1,000 in the suture arms versus 0.7/1,000 in sheds important light on the possible explanations the sutureless securement device arm (P Ͻ .03). A for catheter-related infection, and allowed the meta-analysis of two trials (including the one home care agency to target surveillance activities above) compared stabilization devices with tradi- to high-risk patients. The infrastructure of pro- tional catheter securement procedures with tape spective data collection allows the agency to track or surgical strips and reported a statistically sig- outcomes data and respond promptly to changes. niﬁcant decrease in catheter-related blood stream Another practice inconsistent with the evidence infection rates with the use of stabilization surrounds the techniques used to secure cathe- devices.17 Hypothetically, the differences are be- ters. Historically, catheters have been sutured im- cause of the skin disruption caused by suture mediately after placement, and nurses have relied placement, as well as use of unclean tape applied on mounds of tape to reduce the risk of dislodg- to non-intact skin.18 ment. A trial of 170 peripherally inserted central One institution reported 6-year outcome data catheter (PICC) line recipients, randomized sub- following the implementation of a midline cathe- jects to standard securing procedures with sutures ter placement program for high-risk patients.19 and tape versus a sutureless securement device.16 Midline catheters have longer cannulas than pe- The infection rate per 1,000 catheter days was ripheral catheters (7 to 8 inches v 1 inch, respec- PREVENTION OF INFECTION IN CANCER PATIENTS 179

tively), yet are indicated for shorter infusion time Antibiotic locks are a more recent, controver- than PICCs. The vascular team reviewed the scant sial intervention in vascular catheter care. Solu- literature and, in accordance with expert opin- tions of varying antibiotics and concentrations are ions, developed explicit criteria for midline place- instilled in the catheter for up to 12 hours. Six ment. The program reports increased rates of in- randomized trials were pooled in meta-analysis to fusion completions for patients with midline examine the effect of antibiotic locks on catheter- catheter placement, with low phlebitis and infec- related blood stream infection. The relative risk of tion rates. An important implication of this study catheter-related blood stream infection was signif- is that in the absence of strong evidence, expert icantly decreased in the antibiotic lock arm, com- opinion was consulted to drive interventions. Sub- pared with usual care (P Ͻ .01).22 However, the sequently, the change in practice was studied sys- Healthcare Infection Control Practices Advisory tematically and published in the peer-reviewed Committee (HICPAC) does not recommend anti- literature. biotic locks in routine practice because of resis- For decades, the standard of care for the patient tance concerns.23 Studies using standardized drug requiring bone marrow or stem cell transplanta- and concentration, in immunocompromised pa- tion for hematologic malignancy was placement of tient populations, with detailed antimicrobial re- a tunneled Hickman catheter. However, other de- sistance data are required before adoption of this vices have documented efficacy in this high-risk practice can be widely supported for oncology population. In a single-site, retrospective chart nursing practice. review of 120 autologous marrow/stem cell recipients, the outcomes for those who received Hick- Pneumonia man catheters versus PICCs were compared.20 Health care-associated pneumonia is a frequent While all Hickman catheters were placed by sur- and severe problem for patients. The matrix of geons, the PICCs were placed by nurses and radi- immunosuppression, surgery, nutritional defi- ologists. Cellulitis, phlebitis, and bacteremia were ciencies, and activity intolerance makes pneumo- less frequent in the PICC group. Although the nia particularly problematic for patients with can- groups were relatively well balanced in terms of cer. The CDC has published exhaustive, evidence- age, gender, and presenting diagnosis, comorbid- based guidelines for pneumonia prevention in ity and other uncaptured indicators of illness se- health care facilities.24 To date, this guideline has verity were not available. The data, however, do been slowly adopted. However, a single institution point to encouraging results in terms of reducing has reported recent improvement in nosocomial infection in a vulnerable patient population. pneumonia rates after the implementation of a A recent study examined whether patients with performance improvement project to adhere to tunneled catheters required dressings once cuff the CDC guidelines.25 adherence had occurred.21 The rationale support- The ONS prevention of infection PEP team sum- ing dressing-free catheters include the observa- marized the 2003 recommendations by the CDC tion that dressings may place and contain patho- on the PEP card,6 with additional detail at the gens closer to the catheter site. Second, the website (www.ons.org/outcomes).7 Important Dacron cuff in the tunneled portion of the catheter pearls from this review for the oncology popula- adheres to tissues along the tunnel, preventing tion include: displacement. This randomized, single-site study followed 78 patients with cancer who received 1. Clean, disinfect, rinse, and dry nebulizers be- Quinton or Cook catheters with two or three lu- tween doses. Use sterile medication (single mens. At day 21, afebrile patients with low bacte- doses are preferred) and administer asepti- rial counts on exit site culture were randomized to cally. skin cleansing with chlorhexidine and scheduled 2. Oxygen tubing, humidification circuits and gauze dressings or skin cleansing without dress- masks should be changed after malfunction or ing. While sepsis rates were lower in the no-dress- visible signs of contamination. ing arm, the result did not reach statistical signif- 3. Glove use by providers when handling respira- icance. However, the average days until blood tory secretions; Single-use gowns when respi- stream infection were significantly longer in the ratory secretions are anticipated. no-dressing arm versus scheduled gauze dressing 4. 23-valent pneumococcal polysaccharide vacci- arm (145 v 90 days, P Ͻ .03). nation to adults with cancer. 180 C.R. FRIESE

cols, even after adjustment for stroke severity. TABLE 3. Interventions to Prevent Transmission of Drug- Applied to oncology practice, clinicians should Resistant Organisms32 identify patients at higher risk for dysphagia and arrange for screening. Head, neck, and esophageal Category Specific Interventions patients are candidates, but additional patients may be considered based on comorbidities, and Administrative Organizational priority on infection control. history of surgery and radiotherapy. Tracking system of patients colonized with resistant organisms. Use of coalitions to track and compare data. At Aggressive Organisms least annual feedback to health care providers on infection rates and Three organisms are responsible for the large pro- outcomes portion of infectious diseases difficult to eradicate Education Annual, updated training to health care providers on pathogen incidence, risk with standard therapy: Clostridium difficile-asso- factors, and evidence-based control ciated disease (CDAD), VRE, and MRSA. While efforts clinically heterogeneous diseases, several princi- Antimicrobial Strongly implicated in resistant organisms. ples of care are relevant to all three in terms of use Frequent updates of facility-specific prevention. First, the influence on hand hygiene resistance information, formulary restrictions, interdisciplinary review of cannot be understated. Clear instructions based anti-infective use on empirical science are provided by the CDC’s Surveillance Standardized laboratory methods to test Healthcare Infection Control Practices Advisory for resistant organisms. Rapid Committee27 and are summarized in the ONS PEP notification system to health care website.7 Whereas the CDC guideline details in- providers. Unit-specific resistance reports to target interventions and structions for both water-based and water-less increased surveillance scenarios, it provides the caveat that most alcohol Infection Standard precautions in all health care hand rubs are not effective against spore-based control settings. Contact precautions for organisms, such as CDAD. While some contradic- patients with resistant organisms in tory evidence suggests that use of alcohol rubs acute care settings. Consider increased efforts in long-term care. Standard decreases CDAD rates in hospitals, it is clear that precautions appropriate for home care/ handwashing alone is not sufficient to combat ambulatory settings. For patients with these aggressive pathogens.28,29 Additional inter- resistant organisms, providers wear ventions, such as isolation procedures, are re- masks when splashing is anticipated, quired to prevent transmission of these organisms open tracheostomy care, care of open, heavily colonized sources to patients. Environment Specific cleaning, disinfection, and The CDC has articulated isolation procedures to sterilization procedures. Focus on prevent the transmission of resistant pathogens to frequently touched surfaces other patients. These procedures are published by the evidence-based document endorsed by the HICPAC, and were most recently revised online in April 2005.30,31 The baseline approach, standard 5. Restrictions on visitors when respiratory symp- precautions, provides the initial framework for toms are present. interactions between health care providers and While not addressed by the CDC guideline, a patients; the goal is to reduce the transmission of recent contribution to the literature was the ex- microorganisms between recognized and unrecog- amination of formal dysphagia screening for pa- nized sources in hospitals. The second approach, tients at risk for pneumonia.26 The Stroke Practice termed “transmission-based precautions,” focuses Improvement Network conducted chart review in on interrupting the spread of highly virulent 15 institutions, and tracked adherence to formal pathogens throughout health care facilities. An dysphagia screening protocols. Patients who re- example is the use of contact precautions to received formal dysphagia evaluation after cerebro- duce the risk of CDAD, VRE, and MRSA transmis- vascular event but before the re-initiation of oral sion from colonized patients to other patients. A intake were deemed successfully screened. Rates parallel intervention, empiric precautions, is used of pneumonia were significantly higher in hospi- to identify patients with clinical characteristics tals that lacked formal dysphagia screening proto- who should be isolated (using contact precau- PREVENTION OF INFECTION IN CANCER PATIENTS 181

Common to All Pathogens

Vascular devices Urinary catheters Intubation/mechanical ventilation Surgical procedures (especially organ transplantation or gastrointestinal surgery) Antimicrobial therapy (any type) Severity of illness Prolonged length of stay Transfers within institution

C diff. VRE MRSA

Third-generation cephalosporins Comorbidities Recent hospitalization Clindamycin Hospitalization in past 3 months Chronic illnesses Intravenous Vancomycin Long-term care stay Women Use of enteral feeding tubes Altered mental status Diabetes mellitus Intensive care unit stay Long term care stay in past 2 years

C diff: Clostridium dificile; VRE: Vancomycin-resistant Enterococcus; MRSA: Methicillin-resistant Staphylococcus aureus

FIGURE 1. Risk factors for increased risk of resistant pathogens.28,36-45

tions) before definitive microbiological culture. effective.33 To assist in surveillance activities, recent For example, diarrhea in patients with inconti- studies that identify risk factors for the colonization nence or diaper use, or diarrhea in an adult with of resistant organisms were reviewed and are pre- recent antibiotic use, would raise suspicion for sented in Figure 1. Many risk factors are shared CDAD and warrant empiric precautions until after across pathogens, such as vascular catheters, uri- CDAD toxins return negative. Patients with active, nary catheters, prolonged hospital stay, and trans- infected wounds that cannot be covered com- fers within hospitals. Concurrent antibiotics are also pletely are also candidates for empiric precau- common factors for all three pathogens, but are tions. most strongly implicated in Clostridium difficile in- The recent CDC guideline on the prevention of cidence; cephalosporins and clindamycin are recog- multi-drug-resistant organisms in health care set- nized culprits. Oncology nurses can identify patients tings outlines six initiatives to prevent the transmis- at risk for resistant organisms, conduct active sur- sion of organisms such as VRE and MRSA (see Table veillance in the form of querying for symptoms and 3).32 These initiatives support clinicians in their surveillance culture, and isolate patients at high infection control efforts. When prevalence of resis- suspicion for resistant pathogens. Although such a tant organisms fails to decrease after these initia- targeted approach has not been published in the tives, or a clinically important pathogen is isolated in oncology literature, it applies the research findings a facility, the guideline suggests more aggressive from active surveillance and risk factor analysis. efforts. The evidence base is strongest for acute care Further, it promotes consistency across providers facilities, and weaker in long-term care, ambulatory, and allows for straightforward outcomes assessment. home care, and infusion settings. Research in this The use of intranasal mupirocin has been inves- area is strengthened when outcomes are compared tigated as a strategy to reduce both MRSA and VRE. before and after interventions, baseline infection While data are more supportive for prevention of control measures are clearly described, changes in MRSA in the surgical patient,34 data on non-surgical practice in addition to infection control procedures patients to prevent MRSA are equivocal.35 In addi- is documented, and data are collected across several tion, valid concerns regarding promoting resistance health care facilities. restrict the prophylactic use of intranasal mupirocin The establishment of a surveillance program has to high-risk surgical populations. been implicated in the reduced incidence of CDAD, VRE, and MRSA. Clinical active surveillance has been defined as prospective screening for VRE via CONCLUSION rectal swabs on admission. This approach, con- trasted with laboratory screening (additional culture hile this review highlights many interven- performed for VRE from stools sent for Clostridium W tions oncology nurses can implement to re- difficile toxin assays) is more sensitive and cost duce the burden of infection in patients with can- 182 C.R. FRIESE

cer, additional studies are needed to conﬁrm comes to providers and institutions. For these current ﬁndings and effects of interventions. Nurs- reasons, prevention of infection, whenever possi- ing research in this area would be strengthened by ble, is an important part of oncology nursing prac- standardized measures, in multiple sites, and fo- tice. Several evidence-based interventions have cused on the oncology population. Much of the been discussed here. The role of handwashing is work reviewed here was performed in general central; providers, patients, and caregivers re- hospital or community settings. The unique clin- quire frequent instruction and observation of ical characteristics of patients with cancer and handwashing practices to assure technical accu- their response to nursing interventions to prevent racy. Across the spectrum of pathogens, recogni- infection are of special priority. tion of high-risk pathogens, consistent application Clearly, the problem of infection in the patient of guideline recommendations, and collection of with cancer affects not only immediate clinical outcome data, can all help optimize outcomes for outcomes, but quality of life and adverse out- this vulnerable patient population.31

REFERENCES

1. Friese CR. Measuring oncology nursing-sensitive patient 15. Gorski LA. Central venous access device outcomes in a outcomes: evidence based summary: prevention of infection. homecare agency: a 7-year study. J Infusion Nurs 2004; 2004. Available at: www.ons.org/outcomes/measures (accessed 27:104-111. December 27, 2006). 16. Smith B. Peripheral intravenous catheter dwell times: a 2. Infectious Diseases Society of America. Facts About In- comparison of 3 securement methods for implementation of a fectious Diseases. 2006. Available at: http://www.idsociety.org 96-hour scheduled change protocol. J Infusion Nurs 2006; (accessed January 26, 2006). 29:14-17. 3. Sipsas NV, Bodey GP, Kontoyiannis DP. Perspectives for 17. Schears GJ. Summary of product trials for 10,164 pa- the management of febrile neutropenic patients with cancer in tients: comparing an intravenous stabilizing device to tape. the 21st century. Cancer 2005;103:1103-1113. J Infusion Nurs 2006;29:225-231. 4. Given BA, Sherwood PR. Nursing-sensitive patient out- 18. Frey AM, Schears GJ. Why are we stuck on tape and comes-a white paper. Oncol Nurs Forum 2005;32:773-784. suture? A review of catheter securement devices. J Infusion 5. National Quality Forum. National voluntary consensus Nurs 2006;29:34-38. standards for nursing-sensitive care: an initial performance 19. Anderson NR. Midline catheters: the middle ground of measure set. 2004. Available at: http://www.qualityforum.org/ intravenous therapy administration. J Infusion Nurs 2004; pdf/reports/nsc.pdf (accessed June 22, 2007). 27:313-321. 6. Zitella LJ, Friese CR, Hauser J, et al. Putting evidence into 20. McDiarmid S, Hamelin L, Huebsch LB. Leading change: practice: prevention of infection. Clin J Oncol Nurs 2006; retrospective evaluation of a nurse-led initiative in vascular 10:739-750. access options for autologous stem cell transplant recipients 7. Zitella LJ, Friese CR, Gobel BH, et al. Putting evidence into ranging from Hickman catheters to peripherally inserted cen- practice: Prevention of infection. 2006. Available at: www.ons.org/ tral catheters. J Infusion Nurs 2006;29:81-88. outcomes/volume1/prevention.shtml (accessed June 22, 2007). 21. Olson K, Rennie RP, Hanson J, et al. Evaluation of a 8. Blot F, Nitenberg G. New tools in diagnosing catheter- no-dressing intervention for tunneled central venous catheter related infections. Support Care Cancer 2000;8:287-292. exit sites. J Infusion Nurs 2004;27:37-44. 9. Penwarden LM, Montgomery PG. Developing a protocol 22. Crnich CJ, Maki DG. The promise of novel technology for obtaining blood cultures from central venous catheters and for the prevention of intravascular device-related bloodstream peripheral sites. Clin J Oncol Nurs 2002;6:268-270. infection. II. Long-term devices. Clin Infect Dis 2002;34: 10. Centers for Disease Control and Prevention Hospital 1362-1368. Infections Program. National nosocomial infections surveil- 23. O’Grady NP, Alexander M, Dellinger EP, et al. Guidelines lance system (NNIS) report, data summary from October 1986- for the prevention of intravascular catheter-related infections. April 1996, issued May 1996: a report from the NNIS system. Am J Infect Control 2002;30:476-489. Am J Infect Control 1996;24:380-388. 24. Tablan OC, Anderson JL, Besser R, et al. Guidelines for 11. Gobel BH, Beck SL, O’Leary C. Nursing-sensitive patient preventing health care-associated pneumonia, 2003 recom- outcomes: The development of the putting evidence into prac- mendations of CDC and the Healthcare Infection Control Prac- tice resources for nursing practice. Clin J Oncol Nurs 2006; tices Advisory Committee. MMWR Morb Mortal Wkly Rep 10:621-624. 2003;53(RR03):1-36. 12. Hadorn DC, Baker D, Hodges JS, et al. Rating the quality 25. Keith DD, Garrett KM, Hickox G, et al. Ventilator-asso- of evidence for clinical practice guidelines. J Clin Epidemiol ciated pneumonia: improved clinical outcomes. J Nurs Care 1996;49:749-754. Qual 2004;19:328–325. 13. Camp-Sorrell D, ed. Access device guidelines: recom- 26. Hinchey JA, Shephard T, Furie K, et al. Formal dyspha- mendations for nursing practice and education. Ed 2. Pitts- gia screening protocols prevent pneumonia. Stroke 2005; burgh, PA: Oncology Nursing Society: 2004. 36:1972-1976. 14. Infusion Nurses Society. Infusion nursing standards of 27. Boyce JM, Pittet D. Guideline for hand hygiene in health practice. J Infusion Nurs 2006;29(suppl 1):S1-S92. care settings. Am J Infect Control 2002;30:S1-S46. PREVENTION OF INFECTION IN CANCER PATIENTS 183

28. Sunenshine RH, McDonald LC. Clostridium difficile- 37. Coia JE, Duckworth GJ, Edwards DI, et al. Guidelines for associated disease: new challenges from an established patho- the control and prevention of methicillin-resistant Staphylo- gen. Cleve Clin J Med 2006;73:187-97. coccus aureus (MRSA) in healthcare facilities. J Hosp Infect 29. Boyce JM, Ligi C, Kohan C, et al. Lack of association 2006;63(suppl 1):S1-S44. between the increased incidence of Clostridium difficile-asso- 38. Salgado CD, Farr BM, Calfee DP. Community-acquired ciated disease and the increasing use of alcohol-based hand methicillin-resistant Staphylococcus aureus: a meta-analysis rubs. Infect Control Hosp Epidemiol 2006;27:479-483. of prevalence and risk factors. Clin Infect Dis 2003;36: 30. Garner JS, Hospital Infection Control Practices Advisory 131-139. Committee. Guideline for isolation precautions in hospitals. 39. Palmore TN, Sohn S, Malak SF, et al. Risk factors for Infect Control Hosp Epidemiol 1996;17:53-80. acquisition of Clostridium difficile-associated diarrhea among 31. Garner JS, Hospital Infection Control Practices Advisory outpatients at a cancer hospital. Infect Control Hosp Epidemiol Committee. Guideline for isolation precautions in hospitals. 2005;26:680-684. April 1, 2005. Available at: http://www.cdc.gov/ncidod/dhqp/ 40. Modena S, Bearelly D, Swartz K, et al. Clostridium gl_isolation.html (accessed December 27, 2006). difficile among hospitalized patients receiving antibiotics: a 32. Siegel JD, Rhinehart E, Jackson M, et al, Hospital Infection case-control study. Infect Control Hosp Epidemiol 2005;26: Control Practices Advisory Committee. Management of multi- 685-690. drug-resistant organisms in healthcare settings, 2006. Available 41. Bader MS. Staphylococcus aureus bacteremia in older at: http://www.cdc.gov/ncidod/dhqp/pdf/ar/mdroGuideline2006. adults: predictors of 7-day mortality and infection with a pdf (accessed December 27, 2006). methicillin-resistant strain. Infect Control Hosp Epidemiol 33. Shadel BN, Puzniak LA, Gillespie KN, et al. Surveillance 2006;27:1219-1225. for vancomycin-resistant Enterococci: type, rates, costs, and 42. Asha NJ, Tompkins D, Wilcox MH. Comparative analysis implications. Infect Control Hosp Epidemiol 2006;27: of prevalence, risk factors, and molecular epidemiology of 1068-1075. antibiotic-associated diarrhea due to Clostridium difficile, 34. Young LS, Winston LG. Preoperative use of mupirocin Clostridium perfringens, and Staphylococcus aureus. J Clin for the prevention of healthcare-associated Staphylococcus Microbiol 2006;44:2785-2791. aureus infections: a cost-effectiveness analysis. Infect Control 43. McDonald LC, Owings M, Jernigan DB. Clostridium dif- Hosp Epidemiol 2006;27:1304-1312. ficile infection in patients discharged from US short-stay hos- 35. Wertheim HF, Vos MC, Ott A, et al. Mupirocin prophy- pitals, 1996-2003. Emerg Infect Dis 2006;12:409-415. laxis against nosocomial Staphylococcus aureus infections in 44. Arango JI, Restrepo A, Schneider DL, et al. Incidence of nonsurgical patients: a randomized study. Ann Intern Med Clostridium difficile-associated diarrhea before and after au- 2004;140:419-425. tologous peripheral blood stem cell transplantation for lym- 36. Safdar N, Maki DG. The commonality of risk factors for phoma and multiple myeloma. Bone Marrow Transplant nosocomial colonization and infection with antimicrobial-re- 2006;37:517-521. sistant Staphylococcus aureus, enterococcus, gram-negative 45. Graham PL III, Lin SX, Larson EL. A U.S. population- bacilli, Clostridium difficile, and Candida. Ann Intern Med based survey of Staphylococcus aureus colonization. Ann In- 2002;136:834-844. tern Med 2006;144:318-325.