Table 3

Results from univariable Cox regression of effects of specific variables on postoperative length of stay in a cohort of patients undergoing surgery for cSDH (n=530) pooled across m=40 imputed data sets

HR95% CIP value
Age (per year increase)0.990.99 to 0.990.042*
Sex (male vs female)1.200.99 to 1.450.053*
Referred patient2.201.54 to 3.13<0.001*
 ASA 2 (all vs ASA 1)0.500.31 to 0.820.006*
 ASA 30.340.21 to 0.55<0.001*
 ASA 40.260.15 to 0.45<0.001*
 ASA 50.220.07 to 0.680.009*
mRS 1 (all vs mRS 0)0.860.67 to 1.110.249
 mRS 20.830.58 to 1.180.299
 mRS 30.730.47 to 1.110.142
 mRS 40.680.46 to 0.980.040*
Admission status
 GCS 15 on admission (vs any other)1.111.06 to 1.16<0.001*
 Admission ePOMS (per 1 domain increase)0.910.87 to 0.970.002*
 Cognitive impairment (yes/no)0.570.48 to 0.69<0.001*
 Airways disease0.750.58 to 0.960.020*
 CVS disease0.830.70 to 0.990.040*
 Heart failure0.700.56 to 0.870.001*
 Anticoagulated on admission0.420.35 to 0.50<0.001*
 Creatinine (per 20 µmol/L increase)0.960.92 to 0.990.022*
Day of surgery
 Preoperative deterioration (yes/no)1.070.89 to 1.280.488
 Length of wait (per hour)0.960.92 to 1.010.142
 Time MAP <80 mm Hg (per 10 min increase)0.990.97 to 1.020.720
 Time ETCO2 not 3–5 kPa (per 10 min increase)0.980.92 to 1.050.616
 Fentanyl dose (per 25 µg increase)1.010.99 to 1.040.330
 Volatile maintenance0.990.84 to 1.180.964
 Operative time (per 10 min increase)0.970.95 to 0.990.004*
 Complications (per 1 domain increase in ePOMS)0.570.52 to 0.63<0.001*
  • HRs >1 indicate a more rapid time to postoperative discharge.

  • *Indicate statistical significance at p<0.1, chosen threshold for inclusion in subsequent multivariable modelling. Postoperative complications included as a time-dependent covariate (by day).

  • ASA, American Society of Anesthesiologists Score; CVS, cardiovascular; ePOMS, Electronic Postoperative Morbidity Score; ETCO2, end-tidal carbon dioxide measurement; GCS, Glasgow Coma Scale; MAP, mean arterial pressure; mRS, Modified Rankin Scale.