Incidence of fluoroquinolone-resistant and extended-spectrum β-lactamase-producing Escherichia coli at a comprehensive cancer center in the United States

BACKGROUND

Quinolones are used extensively for prophylaxis in high-risk cancer patients; however, increasing quinolone resistance is being reported. Extended-spectrum β-lactamase (ESBL)-producing E. coli may be associated with increased morbidity and mortality particularly in neutropenic cancer patients.

METHODS

We conducted a retrospective study of consecutive E. coli isolates from January 2009 to August 2009 at our institution. Data on antimicrobial susceptibility of E. coli isolates to commonly used antimicrobial agents and the frequency of ESBL production and fluoroquinolone resistance were gathered based on CLSI guidelines.

RESULTS

There were 443 isolates of E. coli recovered. The majority were from urine cultures (308 isolates, 69.5%). Forty-one (9.2%) isolates were ESBL producing. Nine (18.3%) of the 49 isolates recovered from blood stream infections were ESBL producing. Quinolone resistance was present in 204 isolates (46%). Carbapenems and aminoglycosides retained excellent activity. E. coli resistance to quinolones increased from 13 to 46% in a period of 13 years (p = 0.001).

CONCLUSION

The incidence of resistance to quinolones at our center may be increasing as a consequence of widespread use of quinolones as prophylaxis for neutropenic patients. ESBL-producing E. coli are frequent at our center and are associated with blood stream infections.

Infections in non-myeloablative hematopoietic stem cell transplantation patients with lymphoid malignancies: spectrum of infections, predictors of outcome and proposed guidelines for fungal infection prevention

The overall risk of infections is lower in patients undergoing non-myeloablative allogeneic stem cell transplantation (NST) than in conventional stem cell transplant recipients. We sought to evaluate conditions associated with increased risk of infections after NST. In 81 patients, 187 infection episodes were noted; chronic lymphocytic leukemia (138 episodes/100 person-years) and recipients of matched unrelated donor graft (128 episodes/100 person-years) had higher risk of infection. Only half of the cytomegalovirus (CMV) infections occurred 31-100 days after transplantation. Most patients with CMV infection were non-neutropenic (100%), had lymphoma (76%), were younger (<55 years; 72%) and had received matched related donor (MRD) graft (72%). However, graft-versus-host disease (GVHD) was present in only 15% of these patients. Seven (78%) of nine invasive fungal infections (IFI) were diagnosed >100 days after NST and were associated with high mortality (78%). Most patients with IFI were also not neutropenic (100%), had received MRD graft (100%), had lymphoma (78%) and were given systemic steroids (78%); unlike CMV infection, 67% of these patients also had GVHD. On the basis of our results, we propose that NST recipients with lymphoma treated with high-dose corticosteroids for GVHD be considered for antifungal prophylaxis or pre-emptive antifungal therapy.

Risk factors and attributable mortality of late aspergillosis after T-cell depleted hematopoietic stem cell transplantation

AIM

Invasive aspergillosis occurs in 5-15% of allogeneic hematopoietic stem cell transplant (HSCT) recipients. Through the 1990s there has been an increase in the incidence of late aspergillosis (LA). We report on the incidence, risk factors, and attributable mortality of LA in a cohort of 398 adult and pediatric patients at Memorial Sloan-Kettering Cancer Center from January 1999 through December 2003.

METHODS

LA was defined as occurring > 40 days post HSCT. LA cases were identified by prospective surveillance and examination of a computerized database. Probable or definite aspergillosis was defined by standard EORTC/MSG criteria. Mortality was attributed to LA if it caused or significantly contributed to death.

RESULTS

The overall incidence of LA in our cohort was 4.1%. Median time from stem cell infusion to diagnosis of LA was 164 days (range 68-677) after HSCT. The incidence of LA among unmodified, T-cell depleted, or reduced intensity HSCT was 2.2%, 4%, and 6.8%, respectively (P not significant). Risk factors for LA were grade II-IV acute graft-versus-host disease (GVHD) (P=0.002), chronic GVHD (P=0.01), secondary neutropenia (P=0.02), and reduced intensity conditioning containing alemtuzumab (P=0.01). LA was the immediate cause of death in 1 of 10 (10%) T-cell depleted, 2 of 2 (100%) unmodified, and 1 of 4 (25%) of reduced-intensity HSCT.

CONCLUSIONS

LA developed a median 164 days post HSCT. All-cause 30-day mortality of LA was 56.3%. The majority of LA cases died of concurrent infections and not from invasive aspergillosis.

Work-up for infectious diarrhea after allogeneic hematopoietic stem cell transplantation: single specimen testing results in cost savings without compromising diagnostic yield

BACKGROUND

Diarrhea is a common complication of allogeneic bone marrow transplantation. Microbiologic stool studies are frequently ordered to rule out infectious etiology. The utility of examining multiple stool specimens per diarrheal episode has not been examined.

METHODS

. We performed a retrospective review of 169 adult and pediatric patients who underwent hematopoietic stem cell transplantation at Memorial Sloan-Kettering Cancer Center from January 1, 2000 though December 31, 2001, who had at least 1 microbiologic stool study. We report on the incidence of enteric pathogens in our population and diagnostic yield of stool studies. A diarrheal episode was defined as a 14-day period from the date of the first stool study. Cost savings analysis was based on projected savings from implementation of proposed guidelines to the study population.

RESULTS

A total of 1649 stool tests were performed (mean 10.6 tests per patient). An infectious cause of diarrhea was found in 45 (28.8%) patients. Diagnostic yield was 6.2% for Clostridum difficile toxin assay, 12.9% for viral cultures, and 1.3% for rotavirus enzyme immunoassay. Bacterial cultures for enteric pathogens, examination for parasites, and rotavirus antigen assay combined had 0.5% positive yield.

CONCLUSIONS

Testing of multiple specimens per diarrheal episode did not increase diagnostic yield. The estimated cost savings by implementing single testing for each type of stool study per diarrheal episode was $49,764 annually (in 2001 US dollars). Judicious use of stool tests to evaluate diarrhea results in significant cost savings without compromising diagnostic yield.