clabsi and picc: mpc score 1157
microbiologically confirmed catheter-related bloodstream infection (CRBSI) with a positive catheter tip culture in the setting of suspected PICC infection was evaluated as a sec- ondary analysis.23 Patients in whom CLABSI was suspected but microbiologic findings were not available (eg, PICC removed without confirmatory blood cultures) were excluded.
Statistical Analyses
Putative risk factors associated with CLABSI were assessed according to a previously published and validated conceptual framework for PICC complications.1 In accordance with this framework, covariates associated with CLABSI were sum- marized as patient, provider, and device factors using descriptive variables. Missing data were assumed to be missing at random, with missing values imputed through a 10-fold multiple imputation procedure.24,25 Of all the variables included, only 15 had missing data with an average of 6.4% missing data per missing variable. Patients were censored for CLABSI at the time of PICC removal or at 70 days if the PICC was still in place and no laboratory evidence to suggest CLABSI (ie, positive blood or tip cultures) was present. Unadjusted associations of covariables with time to CLABSI were initially assessed with results expressed as hazard ratios (HRs) and corresponding 95% confidence intervals (CIs). TheMPCscore was developed in accordancewith published
and validated risk assessment tools.26,27 Point values were assigned to create a tool with an easily applied 0 to 3 integer point system, with individual points reflecting the relative magnitude of the regression coefficients in the model. Vali- dation of the final model was performed using methodologic standards for prediction rules.28,29 Baseline covariates with an unadjusted P value ≤.10 in Cox proportional hazards models (with robust sandwich standard error estimates to account for hospital-level clustering) were considered candidate predictors. In keeping with recommended approaches,30 predictors were entered into the multivariable model, with the final model fit by stepwise selection. The proportional hazards assumption was verified, and no covariates violated this assumption. Using coefficients derived from the multi- variable model, integer point values were assigned to each covariable so that a total point score for CLABSI risk could be estimated for each patient. The predictive performance of the MPC score was assessed
using time-dependent AUC values and predicted CLABSI rates at clinically relevant time points over a range of 6 to 40 device days.31 A bootstrap internal validation procedure with 200 bootstrap resamples was performed to determine the optimi- zation of discrimination.32,33 The predictive performance of the model using the original model coefficients and the assigned integer point values was compared to confirm that there was no loss in predictive ability of the model using the integer point system. Significant findings were first reported based on a type I error rate of 0.05 per test, with no correction for multiple testing. To control for false discoveries due to
multiple testing, a Bonferroni correction was next used requiring a P-value threshold of .005 for covariables to remain in the final model.34 All analyses were performed using SAS, version 9.4 software (SAS Institute, Cary, NC) and R version 3.2.4 software (R Foundation for Statistical Computing, Vienna, Austria).
Ethical and Regulatory Oversight
The University of Michigan Medical School’s Institutional Review Board reviewed this study and assigned it a “not regulated” status.
results Patient Characteristics and Outcomes
Over the study period, a total of 23,088 patients received PICCs and were included in this analysis. Within this cohort, a total of 249 (1.1%) patients developed CLABSI according to the modified CDC/NHSN criteria (Figure 1). Patients that developed PICC-CLABSI more often were on hemodialysis (8.0% vs 3.4%; P<.01), were receiving chemotherapy (16.9% vs 2.5%; P<.01), and had past or present hematological malignancy (20.1% vs 2.9%; P<.01) than patients that did not develop this event. Similarly, patients with PICC-CLABSI were also significantly more likely to have experienced an episode of laboratory-confirmed CLABSI in the 3 months prior to PICC insertion (8.8% vs 1.1%; P<.01). Importantly, patients that developed CLABSI also more frequently had another CVC in situ at the time the PICC was placed—often in an ICU setting (29.3% vs 14.1%; P<.01) (Table 1). With respect to provider characteristics, no significant differences in the risk of CLABSI between PICCs placed by vascular access nurses versus interventional radiologists were observed. Similarly, no differences in the number of insertion attempts, arm or vein used for insertion, hospital location for insertion (ie, ICU, general floor, outpatient, or the emergency room) were noted. However, patients that received a PICC to deliver chemotherapy or total parenteral nutrition were more likely to develop a CLABSI than those who received PICCs for intravenous antibiotic therapy (13.3% vs 2.4%, and 10.0% vs 5.1%, respectively; P<.01 for both, respectively). Similarly, longer dwell time (median 15 vs median 11 days; P<.01) and multilumen versus single-lumen devices (78.3% vs 62.6%; P<.01) were more frequently associated with CLABSI.
Patient, Provider, and Device Predictors of PICC-CLABSI
Associations between patient, provider, and device character- istics and PICC-CLABSI are shown in Table 2. Several patient factors such as hemodialysis (hazard ratio [HR], 2.51 [95% CI, 1.38–4.58]), receipt of TPN (HR, 2.14 [95% CI, 1.61–2.85]), and active cancer (HR, 6.37 [95% CI, 4.28–9.48]) were
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124 |
Page 125 |
Page 126 |
Page 127 |
Page 128 |
Page 129 |
Page 130 |
Page 131 |
Page 132 |
Page 133 |
Page 134 |
Page 135 |
Page 136 |
Page 137 |
Page 138 |
Page 139 |
Page 140 |
Page 141 |
Page 142 |
Page 143 |
Page 144
