1 Department of Medical Microbiology, 2 Department of Clinical Epidemiology and Medical Technology Assessment, 3 Department of Internal Medicine, Division of Hematology/Oncology, University Hospital Maastricht, The Netherlands
* Correspondence to: Dr F. H. van Tiel, Department of Medical Microbiology, University Hospital of Maastricht, P. Debyelaan 25, 6202 AZ Maastricht, The Netherlands. Tel: +31-43-3874643; Fax: +31-43-3876643; Email: fvt{at}lmib.azm.nl
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients and methods: The existing literature on PI, and home care of cytopenic patients after chemotherapy, published in the English language, based on a Medline search, is reviewed.
Results: The studies published so far on home care versus hospital care are all non-randomized studies and confirm that home care of cytopenic patients is safe, in terms of morbidity and mortality due to infections. On the other hand, the majority of studies on the comparison of PI with standard hospital care conclude that an infection-preventive effect of PI exists. The pooled statistics performed confirmed that such an effect of PI exists regarding the occurrence of severe infections, although no benefit to mortality has been shown.
Conclusions: Regarding home care, only the results of a prospective, randomized study of sufficient power will enable definitive conclusions to be drawn as to whether home care is equally safe as hospital-based care with PI.
Key words: chemotherapy, cytopenia, home care, infection prevention, protective isolation
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Because of infection-related morbidity, mortality and costs, preventive measures are generally accepted to be a standard of care, and differ between medical centers. Usually, a combination of preventive measures is taken, since it is assumed that each measure contributes to the overall protective effect. Whether this assumption is true or not is a matter of debate; the fact is that since the introduction of these measures more intensive chemotherapy has been applied, without an increase in the infection rate [1].
Although protective (reverse) isolation, to prevent infections, has long been the accepted standard of care for nursing patients with cytopenia after intensive chemotherapy, there has been a recent drive for home-based treatment of these patients. This idea is inspired by the desire to improve the quality of life, but also, possibly more so, by the necessity of cost saving. Over the past years several reports on the treatment of patients after hematopoietic stem cell transplantation (HSCT) have cast doubt on the presumed necessity of PI during leukopenia, in highly susceptible patient categories [914
].
It therefore appears appropriate to critically review the existing literature on home care and compare with that on PI in the hospital care of these patients. This study will focus on (i) the evidence for the safety of home care, compared to hospital care and (ii) the evidence for infection-preventive effectiveness of PI of patients treated in hospital, with, in both cases, colonization and infectious episodes by Gram-negative rods and/or fungi, and mortality as outcome parameters.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The main outcome parameters for this review were colonization with Gram-negative rods and/or fungi, infectious morbidity and mortality.
Statistical analysis
For the statistical analysis of pooled data on morbidity only the randomized studies on PI, which presented their data at a patient level, were included. For each of the outcome parameters in this analysis (Aspergillus infections, major, local fungal infections/candidemia, severe infections/major, local infections and bacteremia/septicemia) the odds ratios (ORs) of the two interventions, i.e. PI and antibiotic prophylaxis (AP), were pooled separately by calculating the weighted average, with weights that are inversely proportional to the variances [15], and tested for heterogeneity. We called this analysis separate model, for which in one study a full factorial design was used [16
]. To pool the OR of one intervention, the two strata, as created by the other intervention, were treated as two different studies. With the separate model it is not possible to use the information from studies in which the study arm and control arm contrast in isolation as well as antibiotic prophylaxis. Therefore, in an additional analysis, ORs were determined using a logistic regression analysis with PI, AP and the categorical variable study center as independent variables. In this combined analysis, which we called combined model, the ORs of both interventions could be determined simultaneously. To summarize the morbidity end points the (log) hazard ratios were pooled [17
].
Definitions
In the four controlled studies on home care, patients were eligible for the home care cohort after stem-cell reinfusion, provided there were no contra-indications [10, 11
, 13
, 14
]. Home care implied discharge to the private home or, as mentioned in the study by Westermann et al., to the private home or to a residential facility, near the hospital, without medical services [10
]. District nurses visited the patients daily. The outpatient cohort was defined as patients discharged to the private home or to a residential facility near the hospital. These patients were seen and treated daily, in the hospital, by transplant physicians and nurses [10
].
Of the studies on home care (Table 3) only one study by Russell et al. defines an infection, based on one or more of the following criteria: fever, a clinical or radiologic focus, or positive cultures [12]. The remaining studies presented their data without defining infection-related morbidity [9
11
, 13
, 14
]. For the end point mortality (Table 3) the follow-up varied from 6 months after entry of the last patient [9
], the aplastic period, defined as the period starting on the day of stem-cell reinfusion and ending with the day of neutrophil recovery to 0.5 x 109/l [10
], not defined and unknown [11
], not defined, but measured as observation time [13
, 14
], to 100 days [12
],
|
|
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
Study design
Second, the study designs differed. In the study of Westermann et al. on 42 patients, allocation and inclusion to the three study groups was performed consecutively: first the inpatient cohort, then the outpatient cohort, and finally the home-care cohort. Herrmann et al. [11] performed a study of 51 patients receiving clinical care in the peri-transplant period at home in comparison to 88 patients with the same diagnoses, cared for as inpatients in a HEPA-filtered bone marrow transplant unit. Allocation to treatment group was based on choice, geographic accessibility, availability of educated care given at home and a clean home environment, and comprehension of the concepts of infection and aseptic techniques. The studies by Svahn et al. [13
, 14
] describe patients treated at home and treated in conventional single rooms with protective isolation.
Colonization data
Only the study by Westermann et al. presents colonization data [10]. In all patients receiving antibiotic prophylaxis (AP), Gram-negative rods were suppressed. However, itraconazole-resistant Candida glabrata and Candida species were isolated from surveillance cultures of stool and throat washings in 50 of 81 courses.
Morbidity data
With regard to morbidity, the data are presented in various ways, precluding a comparison between the studies (Table 3). Westermann et al. [10] present the median number of days with fever >38°C: 3.5 days for home-care patients, 5 days for outpatients and 4 days for the inpatient cohort. The study by Herrmann et al. [11
] presents the percentages of patients with fever >38.5°C, and the mean number of days with fever >38.5°C: a comparable proportion of patients in the home care and hospital cohorts remain free of fever (20/51, 39% versus 37/88, 42%, respectively). Finally, the studies by Svahn et al. [13
, 14
] show data only on the occurrence of bacteremia, suggesting a lower rate of bacteremia for patients treated at home. This difference was deemed statistically not significant in the larger, and more recent, study [14
].
Mortality data
Mortality/survival data for all studies are presented in Table 3 [914
]. In the study by Herrmann et al. no deaths were reported among the control patients, but mortality reached 2/52 for the home care cohort.
Protective isolation
Thirteen studies on PI were found, nine of which were randomized [16, 18
25
]. The total number of included patients was 6438, the number of patients per study ranging from 37 to 5065, with a median of 100. The measures taken in PI, including AP and diet, are summarized in Table 2. Initially, plastic-tent isolators with HEPA-filtered air [1
, 20
, 21
, 23
, 25
27
], and laminar airflow rooms with HEPA filtration of air [16
, 18
, 20
23
, 25
, 28
, 29
] were employed. In later years standard single rooms with HEPA or without HEPA [19
], and laminar airflow rooms without HEPA filtration [30
] were described as protective environments.
Study design
Nine prospective, randomized studies on PI have been published, of which only four allow the comparison of PI and standard ward care, both without additional preventive measures such as AP [16, 19
, 21
, 23
] (Table 1). Of the five remaining randomized studies, three compare patients in PI and on AP with patients on AP only [20
, 22
, 25
], whereas the other studies compare patients in PI and on AP, with patients in standard ward conditions without AP [18
, 24
].
Colonization data
Data on colonization with facultative aerobic Gram-negative rods and yeasts were presented in six studies [18, 19
, 23
, 24
, 27
, 28
], four of which were randomized studies [18
, 19
, 23
, 24
]. Limited to patients without AP, the comparison from patients with and without PI can be derived from the data of only two studies [19
, 23
]. The study of the EORTC working group [23
] indeed measured an enhanced rate of acquisition and colonization of hospital-acquired strains of Gram-negative rods for patients nursed in standard ward care, compared with those in PI.
The mean colonization rate per patient per week was 0.1, 0.41 and 0.6 for patients in PI and on AP, for patients in PI without AP, and for patients in standard ward care, without protective measures, respectively. Buckner et al. [18] report that 12 of 46 patients retained their antibiotics throughout their stay in PI, and that in one of these 12 patients the suppression of all fecal flora was partial, and that in two patients the suppression was absent. In a third randomized study [19
] cultures were positive for Staphylococcus aureus and/or Pseudomonas aeruginosa in 3.4% of the isolated patients and in 4.4% of patients given standard care.
Morbidity data
Two out of the four randomized studies comparing PI with standard ward care, without AP, demonstrate a preventive effect of PI regarding infection-related morbidity [16, 21
] (Table 4), the two other studies fail to show such an effect [19
, 23
]. All remaining randomized studies claim an effect of the preventive measures [18
, 20
, 22
, 24
, 25
] (Table 4). In summary, seven out of nine randomized studies conclude that PI has a infection preventive effect [16
, 18
, 20
22
, 24
, 25
].
As described in Methods, Aspergillus infections, major, local fungal infections/candidemia, severe infections/major, local infections and bacteremia/septicemia were distinguished as separate end points for further analysis. Only one study [19] presented the data specifically for Aspergillus. Since no infections occurred in the study arm with PI, the effect was tested with a Fisher exact test (P=0.5). For the remaining end points the pooled ORs, calculated by weighting with the inverse variance (separate model) and by the logistic regression model (combined model) are shown in Table 5. All tests for heterogeneity of the ORs were negative (all P-values indicated non-significance).
|
Mortality data
The described effects on infectious morbidity did not translate to a difference in mortality. Table 4 demonstrates mortality or survival data for the randomized studies comparing patients in PI versus patients with standard ward care. The heterogeneity in description of the mortality/survival data hindered the extraction of summary statistics to be used for the pooling method, as proposed in the Methods section. Therefore, we only present a descriptive analysis here.
Eleven out of 13 comparative studies on PI included survival/mortality data [16, 18
20
, 22
26
, 30
]. In these studies mortality was presented in different ways, such as 100-day survival, median survival in weeks, 30-day survival, survival on discharge and survival rate. Survival data of the four studies comparing patients in PI with control patients without additional protective measures [16
, 19
, 21
, 23
] showed that the survival of patients in PI did not differ significantly from that of control patients (Table 4). Two studies conclude that the fatality rate from severe infection [25
] or deaths from infection [24
] was statistically reduced among the patients treated in PI.
Sequence of publication
The time sequence of the published reports is depicted in Table 6, according to whether or not home care is equally safe as hospital care or, in the less recent studies, whether a protective effect had been claimed for PI of patients treated in hospital, in comparison with standard ward care.
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
With respect to the safety of home care of cytopenic patients after chemotherapy, four non-randomized, cohort studies describe successful treatment of HSCT recipients at home, but differ in several respects, including study population, and design of the study [10, 11
, 13
, 14
]. These studies confirm that home care of cytopenic patients is safe, with similar morbidity and mortality in comparison with hospital care. However, these studies are non-randomized and therefore prone to selection bias, and they are too small to allow statistically valid conclusions to be drawn with regard to infectious morbidity. Moreover, these four studies also lack the necessary power, due to the low numbers of patients included, and three of the studies did not present colonization data [11
, 13
, 14
].
The recently published guidelines for preventing opportunistic infections among HSCT recipients state that all allogeneic recipients should be placed in rooms with >12 air exchanges/hour and HEPA filters, but also admit that isolation and barrier precautions in HSCT recipients have not been evaluated [31]. In contrast, one report indicates that allogeneic stem-cell recipients can be treated in hospital without the use of protective isolation [10
].
Colonization data would have provided us with valuable information as to whether or not AP is capable of suppressing the pre-existing aerobic Gram-negative flora, as well as preventing acquisition of strains from the environment, e.g. the diet. For future studies we advocate the evaluation of colonization by aerobic Gram-negative rods, yeasts and non-yeast-like fungi, as an additional end point.
Regarding studies on PI of patients treated in hospital, seven out of nine randomized studies concluded that PI has an infection-preventive effect [16, 18
, 20
22
, 24
, 25
]. The pooled statistics performed for this review reveal an effect of PI regarding the occurrence of infections, although no benefit related to mortality was shown.
With respect to mortality, again the differences between the studies in presenting the data are formidable. This fact alone precludes a fair comparison, even when other factors are taken into consideration, such as type of PI, underlying malignant disease, type of chemotherapy, type of HSCT and the use of growth factors. Nevertheless, a clear statistically significant survival benefit has not been demonstrated by the reviewed studies.
Summarizing, the safety of home care, claimed by four recently published comparative studies, appears to contradict the conclusions of earlier studies on PI of cytopenic patients treated in hospital, the majority of which have shown an infection-preventive effect of PI. Therefore, we conclude that only the results of a prospective, randomized study of sufficient power will enable definitive conclusions to be drawn as to whether home care is equally safe as hospital-based care, with PI.
Received for publication January 20, 2004. Revision received September 8, 2004. Accepted for publication September 10, 2004.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. James KW, Jameson B, Kay HEM et al. Some practical aspects of intensive cytotoxic therapy. Lancet 1967; i: 10451049.
3. Schwartz SA, Perry S. Patient protection in cancer chemotherapy. JAMA 1966; 197: 623627.[CrossRef][Medline]
4. Cruciani M, Rampazzo R, Malena M et al. Prophylaxis with fluoroquinolones for bacterial infections in neutropenic patients: a meta-analysis. Clin Infect Dis 1996; 23: 795805.[ISI][Medline]
5. Engels EA, Lou J, Barza M. Efficacy of quinolone prophylaxis in neutropenic cancer patients: a meta-analysis. J Clin Oncol 1998; 16: 11791187.[Abstract]
6. Aker SN, Cheney CL. The use of sterile and low-bacterial diets in ultraisolation environments. J Parenter Enteral Nutr 1983; 7: 390397.[Abstract]
7. Schimpff SC, Young VM, Greene WH et al. Origin of infection in acute non-lymphocytic leukemia. Significance of hospital acquisition of potential pathogens. Ann Intern Med 1972; 77: 707714.[ISI][Medline]
8. Levine AS, Graw RG Jr, Young RC. Management of infections in patients with leukemia and lymphoma: current concepts and experimental approaches. Semin Hematol 1972; 9: 141179.[ISI][Medline]
9. Russell JA, Poon M-C, Jones AR et al. Allogeneic bone-marrow transplantation without protective isolation in adults with malignant disease. Lancet 1992; 339: 3840.[CrossRef][ISI][Medline]
10. Westermann AM, Holtkamp MMJ, Linthorst GAM et al. At home management of aplastic phase following high-dose chemotherapy with stem-cell rescue for hematological and non-hematological malignancies. Ann Oncol 1999; 10: 511517.[Abstract]
11. Herrmann RP, Trent M, Cooney J et al. Infections in patients managed at home during autologous stem cell transplantation for lymphoma and multiple myeloma. Bone Marrow Transplant 1999; 24: 12131217.[CrossRef][ISI][Medline]
12. Russell JA, Chaudhry A, Booth K et al. Early outcomes after allogeneic stem cell transplantation for leukemia and myelodysplasia without protective isolation: a 10-year experience. Biol Bone Marrow Transplant 2000; 6: 109114.
13. Svahn B-M, Bjurman B, Myrbäck KE et al. Is it safe to treat allogeneic stem cell transplant recipients at home during the pancytopenic phase? A pilot trial. Bone Marrow Transplant 2000; 26: 10571060.[CrossRef][ISI][Medline]
14. Svahn B-M, Remberger M, Myrbäck K-E et al. Home care during the pancytopenic phase after allogeneic hematopoietic stem cell transplantation is advantageous compared with hospital care. Blood 2002; 100: 43174324.
15. Rothman KJ. Modern Epidemiology. Boston/Toronto, Canada: Little, Brown and Company 1986: 178.
16. Petersen F, Thornquist M, Buckner C et al. The effects of infection prevention regimens on early infectious complications in marrow transplant patients: a four arm randomized study. Infection 1988; 16: 199208.[ISI][Medline]
17. Parmar MKB, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med 1998; 17: 28152834.[CrossRef][ISI][Medline]
18. Buckner CD, Clift RA, Sanders JE et al. (and the Seattle Marrow Transplant Team). Protective environment for marrow transplant recipientsa prospective study. Ann Intern Med 1978; 89: 893901.[ISI][Medline]
19. Nauseef WM, Maki DG. A study of the value of simple protective isolation in patients with granulocytopenia. N Engl J Med 1981; 304: 448453.[Abstract]
20. Levine AS, Siegel SE, Schreiber AD et al. Protected environments and prophylactic antibioticsA prospective controlled study of their utility in the therapy of acute leukemia. N Engl J Med 1973; 288: 477483.[ISI][Medline]
21. Yates JW, Holland JF. A controlled study of isolation and endogenous microbial suppression in acute myelocytic leukemia patients. Cancer 1973; 32: 14901498.[ISI][Medline]
22. Schimpff SC, Greene WH, Young VM et al. Infection prevention in acute nonlymphocytic leukemialaminar air flow room reverse isolation with oral, nonabsorbable antibiotic prophylaxis. Ann Intern Med 1975; 82: 351358.[ISI][Medline]
23. (EORTC Gnotobiotic Project Group Writing Committee). Dietrich M, Gaus W, Vossen J et al. Protective isolation and antimicrobial decontamination in patients with high susceptibility to infectiona prospective cooperative study of gnotobiotic care in acute leukemia patients. 1: Clinical results. Infection 1977; 5: 107114.[ISI][Medline]
24. Ribas-Mundo M, Granena A, Rozman C. Evaluation of a protective environment in the management of granulocytopenic patients: a comparative study. Cancer 1981; 48: 419424.[ISI][Medline]
25. Rodriguez V, Bodey GP, Freireich EJ et al. Randomized trial of protected environment-prophylactic antibiotics in 145 adults with acute leukemia. Medicine 1978; 57: 253266.[ISI][Medline]
26. Bodey GP, Gehan EA, Freireich EJ, Frei E III. Protected environment-prophylactic antibiotic program in the chemotherapy of acute leukemia. Am J Med Sci 1971; 262: 138151.[ISI][Medline]
27. Klastersky J, Debusscher L, Weerts D, Daneau D. Use of oral antibiotics in protected units environment: clinical effectiveness and role in the emergence of antibiotic-resistant strains. Pathol Biol 1973; 22: 512.[ISI]
28. Bodey GP, Freireich EJ, Frei E III. Studies of patients in a laminar air flow unit. Cancer 1969; 24: 972980.[ISI][Medline]
29. Sherertz RJ, Belani A, Kramer BS et al. Impact of air filtration on nosocomial aspergillus infectionsunique risk of bone marrow transplant recipients. Am J Med 1987; 83: 709718.[ISI][Medline]
30. Passweg JR, Rowlings PA, Atkinson KA et al. Influence of protective isolation on outcome of allogeneic bone marrow transplantation for leukemia. Bone Marrow Transplant 1998; 21: 12311238.[CrossRef][ISI][Medline]
31. CDC, the Infectious Disease Society of America, and the American Society of Blood and Marrow Transplantation: Guidelines for preventing opportunistic infections among hematopoietic stem cell transplant recipients. Morb Mortal Wkly Rep 2000; 49: 1125.[Medline]