Table 4.

Risk factors for major bleeding in all patients by 30 days (8 events)

Univariate oddsPMultivariate oddsP
Patient factors     
 Age (Δ10 y) 1.15 .65   
 Weight (Δ10 kg) 0.83 .37   
 Sex (male vs female) 1.01 .99   
 Ethnicity (white vs nonwhite) 0.62 .67   
 Disease (lymphoma vs myeloma) 1.17 .85   
Treatment factors     
 Conditioning chemo (melphalan) 1.00    
  BEAM 1.29 .78   
  H-TBI/Cytoxan+/− VP-16 0.61 .69   
  Other 1.93 .52   
 Anticoagulation (discontinuation) 1.00    
  LMWH 0.29 .29   
  Unfractionated heparin 1.96 .39   
 Proposed platelet threshold (Δ10 × 103/μL) 0.99 .95   
 Average platelet AUC (Δ10 × 103/μL) 1.01 .97   
 Thrombocytopenia duration (Δ1 d) 1.00 .90   
 Total length of stay (Δ1 d) 1.02 .11   
Preconditioning laboratory tests     
 WBC (Δ1 × 109/L) 1.00 .07   
 Hemoglobin (Δ1 g/dL) 0.47 .02   
 Platelet (Δ10 × 103/μL) 1.00 .35   
 Creatinine (Δ1 mg/dL) 7.92 .05 14.49 .03 
 Bilirubin (Δ1 mg/dL) 12.16 <.01 19.63 <.01 
 LDH (Δ100 U/L) 1.66 .01   
 PT (Δ1 s) 2.48 .01 2.96 <.01 
 PTT (Δ1 s) 1.01 .75   
Univariate oddsPMultivariate oddsP
Patient factors     
 Age (Δ10 y) 1.15 .65   
 Weight (Δ10 kg) 0.83 .37   
 Sex (male vs female) 1.01 .99   
 Ethnicity (white vs nonwhite) 0.62 .67   
 Disease (lymphoma vs myeloma) 1.17 .85   
Treatment factors     
 Conditioning chemo (melphalan) 1.00    
  BEAM 1.29 .78   
  H-TBI/Cytoxan+/− VP-16 0.61 .69   
  Other 1.93 .52   
 Anticoagulation (discontinuation) 1.00    
  LMWH 0.29 .29   
  Unfractionated heparin 1.96 .39   
 Proposed platelet threshold (Δ10 × 103/μL) 0.99 .95   
 Average platelet AUC (Δ10 × 103/μL) 1.01 .97   
 Thrombocytopenia duration (Δ1 d) 1.00 .90   
 Total length of stay (Δ1 d) 1.02 .11   
Preconditioning laboratory tests     
 WBC (Δ1 × 109/L) 1.00 .07   
 Hemoglobin (Δ1 g/dL) 0.47 .02   
 Platelet (Δ10 × 103/μL) 1.00 .35   
 Creatinine (Δ1 mg/dL) 7.92 .05 14.49 .03 
 Bilirubin (Δ1 mg/dL) 12.16 <.01 19.63 <.01 
 LDH (Δ100 U/L) 1.66 .01   
 PT (Δ1 s) 2.48 .01 2.96 <.01 
 PTT (Δ1 s) 1.01 .75   

Univariate logistic regression models were built to test the association between major bleeding and all relevant covariates to determine potential predictors of major bleeding. Multivariate logistic regression model using a stepwise estimation by backward and forward selection was then generated from variables with P < .10 from the univariate model. Preconditioning creatinine, bilirubin, and prothrombin time were independently predictive of major bleeding in the final model.

or Create an Account

Close Modal
Close Modal