Search

Prognostic features of serum galactomannan (GM) remain poorly defined in patients with GM-positive invasive aspergillosis (GPA). We identified 93 patients with proven or probable invasive aspergillosis (IA) and GM values of ≥0.50 from January 2005 to March 2009. We used Cox modeling of time to 6- and 12-week mortality for the GM level at the time of diagnosis (GM 0 ), GM decay in the week following diagnosis in 72 patients with ≥2 GM values, other predictors of mortality, and antifungal use during the week following diagnosis. Six-week mortality was 55% in the whole cohort and 43% in patients with ≥2 GM determinations. The hazard ratio (HR) of GM 0 per unit increase and 1-week GM decay per unit decline per week were 1.25 (95% confidence interval [CI], 1.01 to 1.54; P = 0.04) and 0.78 (95% CI, 0.63 to 0.96; P = 0.02), respectively, adjusting for other predictors of IA mortality; these values remained stable after adjusting for antifungal use and were predictive of all-cause mortality at 12 weeks with similar adjusted HR values. We conclude that the combination of GM 0 and 1-week GM decay is predictive of all-cause mortality in patients with GPA, independent of other traditional risk factors for mortality and antifungal exposure, supporting GM decay as a potential surrogate endpoint for future antifungal therapeutic trials.

Background Diagnosis of invasive fungal disease (IFD) is challenging, and it remains a significant cause of morbidity and mortality in immunocompromised patients. The (1-->3)-beta-D-glucan (BG) assay may be a useful adjunct, but its diagnostic performance is not well characterized. Methods We retrospectively assessed the diagnostic indices of the BG assay in patients at risk of IFD who had a compatible clinical syndrome for the diagnosis of IFD a week after initial BG testing and at the end of the hospitalization associated with the first BG value. Patients with IFD were classified according to current European Organization for Research and Treatment of Cancer-Mycoses Study Group criteria, independent of BG results. Results A total of 1308 BG assays were performed for 871 patients. One hundred twelve proven or probable IFD cases were diagnosed within 1 week after initial testing, and 116 cases were diagnosed by the end of hospitalization. Sensitivity of an initial BG level 80 pg/mL for IFD at 1 week was 0.64 (95% confidence interval [CI], 0.55-0.73), specificity was 0.84 (95% CI, 0.81-0.86), the positive likelihood ratio was 3.93 (95% CI, 2.94-5.26), and the negative likelihood ratio was 0.43 (95% CI, 0.31-0.59). Albumin, intravenous immunoglobulin, and hemodialysis were associated with elevated BG levels in patients without IFD (odds ratio, 4.78; 95% CI, 2.59-8.80). After excluding patients with these factors, specificity and the positive likelihood ratio of an initial BG level 80 pg/mL increased slightly. Empirical systemic antifungal treatment did not reduce overall BG sensitivity. Sensitivity was slightly lower among patients with hematologic malignancy or stem cell transplantation. Consideration of BG results would have increased the diagnostic certainty to probable in 54% of possible IFD cases. Conclusions BG level appears to be a fair diagnostic adjunct for IFD in patients with appropriate pretest probability and a suggestive clinical syndrome, especially when checked serially in patients not receiving factors associated with an elevated BG level in the absence of IFD.

Rationale and Objectives The role of transthoracic needle biopsy (TTNB) in patients with hematologic malignancies, particularly in discriminating between malignant and benign etiologies, has not been well studied. Hence, the diagnostic efficacy and safety of computed tomography–guided TTNB were retrospectively evaluated in this population. Materials and Methods The records of 53 patients with hematologic malignancies who underwent TTNB from August 1, 1999, to July 31, 2007, were reviewed. Potential factors for increased diagnostic yield and risk for complications were collected and analyzed, including the status of neutropenia, thrombocytopenia, chemotherapy, and transplant as well as lesion size and location. Both cytopathologic and microbiologic results were assessed. Results The most common underlying hematologic malignancy was non-Hodgkin lymphoma, in 20 patients (37.7%). Lesions were most frequently located in the left upper lobe (16 [30.2%]); 33 lesions were pleural based (63.5%), and nine had cavitation (17.0%). TTNB established specific diagnoses in 22 patients (41.5%): malignancies was found in 12 (22.6%) and infections in 10 (18.9%). Sensitivity for detecting malignancy was 50.0%, and sensitivity for the detection of a specific infection was 40.0%. There were no false-positive results. Complications occurred in nine patients (17.0%): self-limited small-volume hemoptysis in one patient (1.9%) and pneumothorax in eight patients (15.1%), one requiring chest tube placement. The results of TTNB led to changes in antimicrobial therapy for eight of the 22 patients with specific diagnoses (36.4%). Conclusions TTNB is a safe diagnostic procedure in patients with hematologic malignancies and has the potential of making specific diagnoses with minimal morbidity.

Sirolimus-based immunosuppressive regimens in organ transplantation have been associated with a lower than expected incidence of cytomegalovirus (CMV) disease. Whether sirolimus has a similar effect on CMV reactivation after allogeneic hematopoietic stem cell transplantation (HSCT) is not known. We evaluated 606 patients who underwent HSCT between April 2000 and June 2004 to identify risk factors for CMV reactivation 100 days after transplantation. The cohort included 252 patients who received sirolimus-tacrolimus for graft-versus-host disease (GVHD) prophylaxis; the rest received non–sirolimus-based regimens. An initial positive CMV DNA hybrid capture assay was observed in 225 patients (37.1%) at a median 39 days after HSCT for an incidence rate of 0.50 cases/100 patient-days (95% confidence interval [CI], 0.44-0.57). Multivariable Cox modeling adjusting for CMV donor-recipient serostatus pairs, incident acute GVHD, as well as other important covariates, confirmed a significant reduction in CMV reactivation associated with sirolimus-tacrolimus–based GVHD prophylaxis, with an adjusted HR of 0.46 (95% CI, 0.27-0.78; P = .004). The adjusted HR was 0.22 (95% CI, 0.09-0.55; P = .001) when persistent CMV viremia was modeled. Tacrolimus use without sirolimus was not significantly protective in either model (adjusted HR, 0.66; P = .14 and P = .35, respectively). The protective effect of sirolimus-containing GVHD prophylaxis regimens on CMV reactivation should be confirmed in randomized trials.

87 Background: Quality Oncology Practice Initiative (QOPI) retrospectively collects process measures at the practice level. Sample size is determined by number of full-time equivalent physicians. Although intended as an improvement program, there is growing pressure to use QOPI for accountability. Methods: We investigated QOPI sampling strategy for EOL measures in a large three-site community oncology practice for all patients meeting QOPI sampling criteria. Analysis period was July 1 to December 31, 2012. We included all patients submitted for the QOPI EOL module and compared the QOPI sample to the full but non-QOPI cohort using Fisher’s exact test. To assess relative precision, we calculated rates and 95% confidence intervals (CI) for the sampled vs. full cohorts (inclusive of QOPI cohort). Results: 246 deaths were analyzed, with 34 (13.8%) from the QOPI sample. No statistically significant differences were found between sampled and full non-QOPI cohorts. However, sampling was associated with wide CIs compared to precision using the full cohort (Table); e.g. #38 yields a CI width of 24.2% QOPI vs. 4.9% full. Similarly, #47 shows CI width of 27.7% vs. 9.0%. Low statistical significance yet wide CIs indicate comparison data are representative but with variability in precision. Conclusions: Although no statistically significant differences were identified, comparing QOPI sampling strategy to full cohort yielded presence of marked differences in precision. We suggest caution in use of current QOPI EOL sampling strategy if used for purposes other than improvement. [Table: see text]