Patient density, nurse-to-patient ratio and nosocomial infection risk in a pediatric cardiac intensive care unit.

BACKGROUND: An investigation of a Serratia marcescens outbreak in a pediatric cardiac intensive care unit (CICU) suggested that understaffing or overcrowding might have been underlying risk factors. OBJECTIVE: To assess the effect of fluctuations in CICU nurse staffing levels and patient census on CICU nosocomial infection rate (NIR). METHODS: The monthly CICU nursing hours, patient days and nosocomial infections were obtained from retrospective review of administrative, patient and microbiology records during December, 1994, through December, 1995 (study period). The NIR and nursing hours:patient day ratio were then calculated. The correlations between NIR vs. nursing hours, patient days and nursing hours:patient day ratio were determined. RESULTS: The median monthly CICU NIR was 6.9 (range, 0 to 15.2) infections per 1000 patient days; the median number of hours worked per month by CICU registered nurses was 7754 (range, 7133 to 8452) hours; the median number of patient days treated per month was 507 (range, 381 to 590) patient days; and the median monthly nursing hours:patient day ratio was 15.2:1 (range, 13.2:1 to 19.9:1). The strongest linear correlation was observed between the monthly NIR and patient days (r = 0.89, P = 0.0001). There was an inverse correlation between the monthly NIR and nursing hours:patient day ratio (r = -0.77, P = 0.003). CONCLUSIONS: The NIR was most strongly correlated with patient census but also was strongly associated with the nursing hours:patient day ratio. These factors may influence the infection rate because of breaks in health care worker aseptic technique or decreased hand washing. Increased patient census alone may increase the risk of cross-transmission of nosocomial infections. As hospitals proceed with cost containment efforts the effect of fluctuations in patient census and nurse staffing on patient outcomes needs evaluation.

Antimicrobial use and resistance in eight US hospitals: complexities of analysis and modeling. Intensive Care Antimicrobial Resistance Epidemiology Project and National Nosocomial Infections Surveillance System Hospitals.

OBJECTIVE: To evaluate the relation between antimicrobial use and resistance in intensive-care unit (ICU) and non-ICU inpatient areas in eight US hospitals. METHODS: We determined antimicrobial use in terms of defined daily doses, antimicrobial-use density (defined daily doses/1,000 patient days), and percentage resistance for five antimicrobial-organism combinations in the ICU and non-ICU inpatient areas of eight US hospitals participating in project Intensive Care Antimicrobial Resistance Epidemiology. RESULTS: Antimicrobial resistance and use varied tremendously among the eight hospitals. Antimicrobial resistance among these five nosocomial pathogens was significantly higher within the inpatient setting of these hospitals, compared with the outpatient setting. One hospital consistently ranked highest for use of all classes of antimicrobials examined. High antimicrobial use was not associated necessarily with high resistance for a particular antimicrobial-organism pair. CONCLUSION: Antimicrobial use varied significantly across these hospitals, but generally was higher in ICUs. These results suggest that concomitant surveillance of both antimicrobial resistance and antimicrobial use is helpful in interpreting antimicrobial resistance in a hospital or ICU and that further analysis is required to determine the role of variables other than antimicrobial use in a statistical model for predicting antimicrobial resistance.

Serratia marcescens outbreak associated with extrinsic contamination of 1% chlorxylenol soap.

OBJECTIVES: To determine risk factors for Serratia marcescens infection or colonization, and to identify the source of the pathogen and factors facilitating its persistence in a neonatal intensive-care unit (NICU) during an outbreak. DESIGN: Retrospective case-control study; review of NICU infection control policies, soap use, and handwashing practices among healthcare workers (HCWs); and selected environmental cultures. SETTING: A university-affiliated tertiary-care hospital NICU. PATIENTS: All NICU infants with at least one positive culture for S marcescens during August 1994 to October 1995. Infants who did not develop S marcescens infection or colonization were selected randomly as controls. RESULTS: Thirty-two patients met the case definition. On multivariate analysis, independent risk factors for S marcescens infection or colonization were having very low birth weight (< 1,500 g), a patent ductus arteriosus, a mother with chorioamnionitis, or exposure to a single HCW. During January to July 1995, NICU HCWs carried their own bottles of 1% chlorxylenol soap, which often were left standing inverted in the NICU sink and work areas. Cultures of 16 (31%) of 52 samples of soap and 1 (8%) of 13 sinks yielded S marcescens. The 16 samples of soap all came from opened 4-oz bottles carried by HCWs. DNA banding patterns of case infant, HCW soap bottle, and sink isolates were identical. CONCLUSIONS: Extrinsically contaminated soap contributed to an outbreak of S marcescens infection. Very-low-birth-weight infants with multiple invasive procedures and exposures to certain HCWs were at greatest risk of S marcescens infection or colonization.

Serratia marcescens transmission in a pediatric intensive care unit: a multifactorial occurrence.

BACKGROUND: Fourteen patients in the pediatric cardiac intensive care unit (CICU) had > or =1 positive culture for a single strain of Serratia marcescens from April through December 1995 (study period). OBJECTIVES: To identify risk factors for S marcescens infection or colonization in a pediatric CICU. METHODS: Retrospective case-control study. Assessment of CICU infection control practices and patient exposure to CICU health care workers (HCWs). Epidemiologic-directed cultures of the environment and HCWs' hands were obtained. SETTING: Pediatric CICU. PATIENTS: Fourteen patients in the pediatric CICU had > or =1 positive culture for a single strain of S marcescens from April through December 1995 (study period). CICU patients who did not have S marcescens infection or colonization during the study period were randomly selected as controls. RESULTS: A case patient was more likely than a noncase patient to have exposure to a single HCW (odds ratio [OR], 19.5; 95% CI, 2.6-416; P<.003); however, this association was not adequately explained by epidemiologic or microbiologic studies. Interviews suggested that during the outbreak period, handwashing frequency among HCWs might have been reduced because of severe hand dermatitis. CONCLUSIONS: A combination of factors, including breaks in aseptic technique, reduced frequency of handwashing among HCWs before and between caring for patients, decreased attention to infection control practices, and environmental contamination may have indirectly contributed to this S marcescens infections outbreak.

Hospital-acquired infections in the United States. The importance of interhospital comparisons.

To use infection rates as a basis for measuring quality of care, the rates must be meaningful for interhospital comparison. A crude, overall nosocomial infection rate of a hospital provides no means of adjustment for patients' intrinsic or extrinsic risks. Before interhospital comparison, rates should be adjusted for nosocomial infection risk factors. Interhospital comparison of rates requires that a hospital participate in a multicenter surveillance system or aggregated national database. This article outlines a series of questions for hospital administrations to pose before entering such an endeavor.

Detection of bloodstream pathogens in a bacille Calmette-Guérin (BCG)-vaccinated pediatric population in Malawi: a pilot study.

Children in Malawi receive bacille Calmette-Guérin (BCG) vaccination within the first 3 days of life. Thus, we hypothesized that Malawian children infected with the human immunodeficiency type 1 virus (HIV-1) might be particularly vulnerable to dissemination of the BCG Mycobacterium bovis strain with which they were vaccinated. Following informed consent by parents, we studied children admitted to a Malawi general hospital during the 1998 wet and dry seasons. Blood from cohorts of acutely ill children was cultured for bacteria, including mycobacteria, and fungi, and tested for anti-HIV-1 antibodies. It was shown that non-typhi Salmonella and Escherichia coli were the predominant bloodstream pathogens during the wet and dry seasons, and that bloodstream dissemination of the BCG M. bovis strain is uncommon in HIV-1-infected children who receive the BCG vaccine.

Tropical diabetic hand syndrome: risk factors in an adult diabetes population.

OBJECTIVES: To determine risk factors for the tropical diabetic hand syndrome, a condition associated with significant morbidity and mortality in Africa. METHODS: This was a case-control study of a Tanzanian diabetes population presenting with the syndrome during February 1998 to March 2000. A case patient was defined as any patient with diabetes presenting with hand cellulitis, ulceration, or gangrene. Control patients were randomly selected patients with diabetes who had no hand symptoms. RESULTS: Thirty-one case patients and 96 control patients were identified. The median age of case patients was 52 years (range, 28--76 y); 58% were male; 4 patients (16%) died. Precipitating events included papule (n = 6), insect bites (n = 6), boils (n = 5), burns (n = 2), or trauma (n = 3). Case and control patients were similar for presence of micro- and macrovascular disease and occupation. On logistic regression analysis, independent risk factors were body mass index of 20 or less (odds ratio [OR] = 18.0; 95% confidence interval [CI] = 4.3--97.0; P < 0.001), peripheral neuropathy (OR = 23.0; 95% CI = 5.3--124.0; P < 0.001), or type I diabetes, (OR = 6.7; 95% CI = 2.0--24.0, P < 0.01). CONCLUSION: The major risk factors for the tropical diabetic hand syndrome are intrinsically related to the underlying disease. Thus, prevention of hand infections may require aggressive glucose control, and education on hand care and the importance of seeing a doctor promptly at the onset of symptoms.

Seasonal variation in the etiology of bloodstream infections in a febrile inpatient population in a developing country.

OBJECTIVES: Published data suggest that Streptococcus pneumoniae, non-typhi Salmonella species, and Mycobacterium tuberculosis are the predominant causes of bloodstream infection (BSI) in hospitalized populations in sub-Saharan Africa. This study was conducted during the wet season to ascertain the etiology and prevalence of BSI among febrile inpatients in a hospital where the dry season BSI profile in a similar study population had already been documented. METHODS: In the period from March to May 1998, consecutive febrile (> or = 37.5 degrees C) adult (> or = 14 y) patients presenting to a Malawi hospital were enrolled after providing informed consent. Following clinical evaluation, blood was drawn for culture (bacteria, mycobacteria, and fungi), human immunodeficiency virus (HIV) testing, and malaria smears. RESULTS: Of 238 enrolled patients, 173 (73%) were HIV-positive and 67 (28%) had BSI. The predominant wet season BSI pathogens were non-typhi Salmonella species (41%), M. tuberculosis (19%), and Cryptococcus neoformans (9%) (cf. the predominant dry season pathogen was S. pneumoniae). Mycobacteremia was more likely in HIV-positive than in HIV-negative patients (13/173 vs. 0/65; P < 0.05). A logistic regression model yielded clinical predictors of BSI that included chronic fever, oral candidiasis, or acute diarrhea. CONCLUSION: Pathogens causing BSI in febrile inpatients in a Malawi teaching hospital vary by season. Season- and country-specific studies, such as this one, provide data that may facilitate empirical therapy of febrile illnesses whose etiologies vary by season.

Rodent bites on the feet of diabetes patients in Tanzania.

AIM: To characterize the epidemiology of rodent bite foot injuries in patients presenting to a diabetes clinic in Tanzania. METHODS: During July 1998-September 2003, all adult diabetes patients presenting with rodent bite injuries were identified. Follow-up included antimicrobial therapy and surgery, where appropriate. RESULTS: All 34 patients with rodent bites had Type 2 diabetes and peripheral neuropathy. Median age was 55.5 years; 62% were male. All bites occurred during sleep. The median time between acquiring the bite and presentation to MNH was 7 (range: 1-17) days. Patients who delayed seeking medical attention were significantly more likely to develop gangrene. Seventeen patients underwent minor or major amputation. Complete healing occurred in 30 (88%) patients; four patients died. CONCLUSION: Diabetes patients with peripheral neuropathy are at increased risk of bite injuries in areas with large rodent populations. Preventive efforts should include covering the feet at bedtime, and daily feet examination by patient or relatives.

Clinical microbiology in developing countries.

We review the problem of limited microbiology resources in developing countries. We then demonstrate the feasibility of a cohort-based approach to integrate microbiology, epidemiology, and clinical medicine to survey emerging infections in these countries.

Long-term sequelae of Rocky Mountain spotted fever.

Twenty-five patients with definite or probable Rocky Mountain spotted fever (RMSF) who were hospitalized for > or = 2 weeks were identified from our database of 105 patients. Follow-up information was collected for 20 patients, per telephone and/or medical records. The remaining five patients were lost to follow-up or died. Nine patients had > or = 1 long-term sequelae (defined as complications related to an original acute infection with Rickettsia rickettsii that persisted for > or = 1 year following hospital discharge). The ages of patients with sequelae ranged from 2 to 74 years (mean and median, 38 years); duration of follow-up ranged from 1 to 18 years (mean, 11 years). The mean lengths of hospitalization for patients with and without long-term sequelae were 47 days and 20 days, respectively (P < .05). Long-term neurological sequelae included paraparesis; hearing loss; peripheral neuropathy; bladder and bowel incontinence; cerebellar, vestibular, and motor dysfunction; and language disorders. Nonneurological sequelae consisted of disability from limb amputation and scrotal pain following cutaneous necrosis. These data suggest that significant long-term morbidity is common in patients with severe illness due to RMSF.

Fatal hand sepsis in Tanzanian diabetic patients.

Hand infections are common presentations among diabetic patients admitted to hospital in Tanzania. The morbidity and mortality are high and patients' hospital inpatient stay tend to be prolonged because of suboptimal therapy. We describe four diabetic patients with hand infections and fatal outcomes. In contrast to patients with foot infections, none of our patients had clinical evidence of peripheral neuropathy or vascular disease. All four patients eventually died in hospital after acquiring hand sepsis and diabetic ketoacidosis which did not respond to prolonged courses of intravenous insulin and antimicrobials. Literature review suggests such infections are at least as likely to include Gram-negative organisms as Staphylococcus aureus. Primary management should have included aggressive surgery with limb amputation if necessary with adjunctive antimicrobial therapy.

The epidemiology of diabetic limb sepsis: an African perspective.

We review the epidemiology of foot and hand sepsis in adult diabetes patients in Africa. Limb sepsis in these patients is associated with significant morbidity and mortality. The pathogenesis of diabetic foot infections in these patient populations appears to be similar to that for patients in industrialized countries -ulcers and underlying peripheral neuropathy being the most important risk factors. Prevention of peripheral neuropathy through aggressive glycaemic control may be the most important primary control measure for foot infections. The tropical diabetic hand syndrome (TDHS) is being increasingly seen in diabetes patients in certain parts of Africa. The syndrome is acute, usually follows minor trauma to the hand, and is associated with a progressive synergistic form of gangrene. The major risk factors for TDHS are unknown but recent data suggest poor glycaemic control is associated with poor outcome. Treatment of TDHS requires aggressive surgery. Hence, preventive efforts for both foot and hand sepsis include aggressive glucose control, and education on hand and foot care and the importance of seeking medical attention promptly at the earliest onset of symptoms

Clinical outcome of diabetes patients hospitalized with foot ulcers, Dar es Salaam, Tanzania.

AIMS: To determine the prevalence rate, clinical features, risk factors, and clinical outcome of foot ulcers in diabetes patients admitted to Muhimbili National Hospital, Dar es Salaam, Tanzania. METHODS: A prospective cohort study of newly hospitalized, adult diabetes patients with foot ulcers was conducted during January 1997 to December 1998 (study period). Detailed clinical and epidemiological data were recorded for each patient, followed by a comprehensive physical examination. Clinical outcome was documented. RESULTS: Of 627 diabetes patients evaluated during the study period, 92 (15%) had foot ulcers. Of these 92 patients, 30 (33%) were selected for surgery (minor and major amputations); the rest were managed conservatively. Patients who underwent surgery were more likely than those who did not to have gangrene (P < 0.001) or neuropathy (P < 0.01). On stratification by severity of ulcers, patients with Wagner score > or = 4 were significantly more likely than those < 4 to have neuroischaemic foot lesions (P < 0.001) or delayed presentation to hospital (P < 0.001). The overall mortality rates for amputees and non-amputees were similar (29%); the highest in-patient mortality rate (54%) was observed among patients with severe (Wagner grade > or = 4) ulcers who did not undergo surgery. CONCLUSIONS: Diabetic foot ulcers are associated with significant morbidity and mortality in Tanzania. Mortality rates among patients with severe ulcers remain high despite surgery. Thus, surgery undertaken during the less severe stages of ulcers may improve patient outcome. Education of patients should underscore the importance of foot care and consulting a doctor during the early stages of foot ulcer disease.

Serious hand sepsis and diabetes mellitus: specific tropical syndrome with western counterparts.

Infection in the extremities of diabetic patients most commonly involves the feet and, at least in western societies, is often associated with chronic complications of diabetes. Severe hand infection, often culminating in amputation and even death, is, however, well-described in tropical countries, where it may not be associated with any evidence of neuropathy or arterial insufficiency. Similar cases are described in the western literature but are more often associated with more severe antecedent trauma. The literature describing hand sepsis in diabetic patients both in tropical and in western practice is reviewed and we draw some conclusions about pathogenesis and treatment from the literature and from original data documenting the varying experience of hand sepsis in diabetic practice throughout Africa.

Fatal Mycobacterium tuberculosis bloodstream infections in febrile hospitalized adults in Dar es Salaam, Tanzania.

Causes of community-acquired bloodstream infections (BSIs) in sub-Saharan Africa are unknown with regard to mycobacteria and fungi. We prospectively studied 517 consecutive febrile (axillary temperature, > or =37.5 degrees C) adults (> or =15 years of age) admitted to one hospital in Tanzania. After hospital admission and informed consent, blood was drawn for culture (of bacteria, mycobacteria, and fungi), determination of human immunodeficiency virus type 1 (HIV-1) status, and malaria smears. Malaria smears were prepared for a control group of 150 afebrile patients. One hundred and forty-five patients (28%) had BSI. Of these 145 patients, 118 (81%) were HIV-1-infected. HIV-positive patients were more likely than HIV-negative ones to have BSI (118 of 282 vs. 27 of 235; P < .0001). The three most frequently isolated pathogens were Mycobacterium tuberculosis (60 [39%]), non-typhi Salmonella species (29 [19%]), and Staphylococcus aureus (13 [8.3%]). The incidence of malaria parasitemia was similar in study and control patients (9.5% vs. 8%). In this patient population with high prevalence of HIV-1 infection, M. tuberculosis has become the foremost cause of documented BSI.

Albendazole is effective treatment for chronic strongyloidiasis.

A total of 301 British ex-Far East prisoners of war, many of whom worked on the Thai-Burma Railway during World War II, consecutively attended The Liverpool School of Tropical Medicine for clinical review between January 1987 and August 1990. Fifty-two (17%) were found to have chronic strongyloidiasis. Diagnostic criteria included any of the following: characteristic larva currens rash, positive Strongyloides serology, and positive stool examination. Forty-seven were evaluable 6 months, after therapy with albendazole 400 mg twice daily for 3 days, which resulted in a 75% cure rate. Cure was defined as disappearance of the rash, if present, negative serology and negative stool examination. Patients who had more than one positive diagnostic feature were only considered cured if both or all had disappeared or become negative. All 12 of the patients in whom initial treatment failed were given a second course of albendazole: three further cures were obtained in eight evaluable patients. The overall cure rate was 81%. The only side-effects recorded were mild nausea and diarrhoea in one patient. We suggest that albendazole should be the treatment of choice for chronic strongyloidiasis.

Enterobacter cloacae and Pseudomonas aeruginosa polymicrobial bloodstream infections traced to extrinsic contamination of a dextrose multidose vial.

OBJECTIVE: To identify risk factors for polymicrobial bloodstream infections (BSIs) in neonatal intensive care unit (NICU) patients during an outbreak of BSIs. DESIGN: During an outbreak of BSIs, we conducted a retrospective cohort study, assessed NICU infection control practices and patient exposure to NICU healthcare workers (HCWs), and obtained cultures of the environment and HCW hands. PATIENTS: During the period May 3 to 7, 1996, 5 infants contracted BSIs caused by both Enterobacter cloacae and Pseudomonas aeruginosa, and one infant contracted a BSI caused by E cloacae only. For each pathogen, all isolates were identical on DNA typing. RESULTS: Infants exposed to the following were more likely than nonexposed infants to have BSI: umbilical venous catheters (6/14 vs 0/7, P = .05), total parenteral nutrition given simultaneously with a dextrose/electrolyte solution (6/12 vs 0/9, P = .02), or one HCW (5/7 vs 1/13, P = .007). Neither environmental nor HCW hand cultures yielded the outbreak pathogens. Quality control cultures of intravenous solution bags were negative. CONCLUSIONS: We speculate that a dextrose multidose vial became contaminated during manipulation or needle puncture and that successive use of this contaminated vial for multiple patients may have been responsible for BSIs. Aseptic techniques must be employed when multidose vial medications are used. Single-dose vials should be used for parenteral additives whenever possible to reduce the risk of extrinsic contamination and subsequent transmission of nosocomial pathogens.

Age-related differences in cell-specific cytokine production by acutely ill Malawian patients.

Age-related changes in human cell-specific cytokine responses to acute illness have not been well examined. We therefore evaluated age-related differences in T, B and natural killer (NK) peripheral blood lymphocyte cytokine responses of 309 acutely ill hospitalized people in Malawi, Africa, < 1 month-61 years of age. We used four-colour flow cytometry and performed Wilcoxon rank sum and Kruskal-Wallis tests, Pearson (rp) and Spearman (rs) correlations, and linear and logistic regression analyses to control for human immunodeficiency virus infection (HIV) status, the percentages of lymphocytes expressing CD4, and the nature of the acute infection. The percentages of CD8- and CD8+ T cells producing induced IL-8 decreased with age (rs = -0.44 and -0.53). The percentages of T cells producing TNF-alpha were higher, and the percentages producing IL-10 were lower, in those > or =13 than those < 13 years old (medians: 17.7 versus 10.5 and 1.4 versus 3.0, respectively). The percentages of CD8- T cells producing IFN-gamma were higher and stable in those > or =1 year old compared to infants (medians: 23.5 versus 10.4); the percentages of NK producing IFN-gamma were higher post-infancy and then declined to relatively low levels with increasing age. The percentages of T cells producing IL-2 were highest in those 5- <31 years old (median 5.6) and lowest in those > or =31 years old (median 1.9). The ratios of the percentages of T cells producing IL-4 to those producing IL-8 and to those producing IL-10 both increased with age. These data suggest that innate immunity, represented by NK IFN-gamma production, dominates in early life. A number of shifts occur after infancy and before adolescence, including a proinflammatory shift from IL-8 to TNF-gamma and a type 2 shift from IL-10 to IL-4 dominance. These findings suggest distinct age-related differences in the human response to acute illness and may be useful in directing future efforts at immunomodulatory therapies.