Patient Care in Rapid-expansion Intensive Care Units During the COVID-19 Pandemic Crisis

Jade I. Basem; Anna F. Roth; Robert S. White; Virginia E. Tangel; Silis Y. Jiang; Jacky M. Choi; Katherine L. Hoffman; Edward J. Schenck; Zachary A. Turnbull; Kane O. Pryor; Natalia S. Ivascu; Stavros G. Memtsoudis; Peter A. Goldstein


BMC Anesthesiol. 2022;22(209) 

In This Article


From March 3 to May 19, 2020, 343 patients were admitted into traditional ICUs and 68 were admitted to an Expansion-ICU. Reason for admission to any ICU at NYP-WCMC was COVID-19-related acute hypoxemic respiratory failure requiring mechanical ventilation. Of the 68 patients admitted to an Expansion-ICU, there was incomplete information on 2 patients, resulting in a final sample size of 66. Of these 66 patients, 60 were also in a standard ICU sometime during their hospital stay. Table 1 displays the characteristics of study cohort, overall and comparing the Expansion-ICU cohort to the Referent cohort. For the Expansion-ICU, there were 45 male (68%) and 21 (32%) female patients; the median age (years; IQR) was 62 (51–70). The majority of patients self-identified as White (32%) or Latino (39%); 7 records (11.0%) did not indicate race or ethnicity. Geographically, patients were primarily from the boroughs of Brooklyn (32%), Manhattan (19%), or Queens (19%). Diagnosed comorbidities prior to admission included 37 with hypertension (56%), 20 with diabetes (30%), and 17 with pulmonary disease (26%). Among Expansion-ICU patients, 33 (50%) were obese and 21 (32%) were overweight. Non-smokers made up the majority of the study group (82%). There were no significant differences in baseline covariates between patients admitted to an Expansion-ICU compared to those never admitted to an Expansion-ICU.

Laboratory results were collected when available for patients 24 h upon first admission to standard ICU (for referent cohort) or admission to Expansion-ICU (Table 2). Upon first ICU admission, the Expansion-ICU cohort had significantly lower levels of ferritin (1009 [530,1464] vs.1446 [87,2000], q = 0.022), lactate dehydrogenase (LDH) (476 [356,576] vs 572 [456,801], q = 0.007), and WBC count (9.1 [6.2, 11.5] vs 11.1 [7.4, 15.1], q = 0.022); in contrast, lymphocyte counts were higher in the Expansion cohort [median (IQR): 10 (5, 15), n = 48 vs 7 (4, 12), n = 246 in the Referent cohort, P = 0.045]. In Expansion ICU patients, the median D-Dimer level was 1,098 (IQR: 631, 2,805; n = 24), which was not significantly different (P = 0.4) from that observed in the Referent ICU population [median (IQR): 1,749 (641, 4382); n = 158].

There were statistically significant differences for time to intubation, hospital LoS, ICU LoS, and time on a ventilator between referent and Expansion cohorts (Table 3). Among discharged patients, the referent cohort had higher median time to intubation (1.88 [0.66, 4.90] vs 1.14 [0.51, 2.80] days, q = 0.039), but lower median time than the Expansion cohort for hospital LoS (40 [22,62] vs 49 [40,69], q = 0.002), ICU LoS (17 [9,30] vs 32 [25,50] days, q < 0.001), and length of time on a ventilator (17 [10,33] 33 [23,34], q < 0.001). Of note, time to intubation was calculated from admission to the emergency department (ED; or transfer in date) to intubation; however, there were many patients that were intubated before ED admission or transfer in date, giving a negative time to intubation—these patients were not included. Consequently, this variable is limited to patients with a positive time to intubation. For the patients who died, the only significant difference between the groups was that the Expansion cohort had a higher median ICU LoS (25 [16,26] vs. 12 [7,20] days, q = 0.037) and ventilatory length (30 [19,32] vs. 13 [5,20] days, q = 0.015).

Management aspects of patients during their ICU stay is in Table 4. The use of vasopressors and prone positioning between the two cohorts was similar regardless of patient disposition. For patients that were discharged, the Expansion-ICU cohort had statistically significantly higher rates of use of propofol (85% vs 67%, q = 0.025), dexmedetomidine (83% vs 42%, q < 0.001), and ketamine (11% vs 2.3%, q = 0.025). For patients who died, there were no statistically significant differences in treatment.

Disposition data was collected for all patients (Table 5). For the Expansion-ICU cohort, 17 (26%) were discharged home, 31 (47%) were discharged to a subacute rehabilitation facility, and 12 (18%) died in the hospital. The two cohorts had similar rates of discharge to home, but the Expansion cohort had higher rates of discharge to a rehabilitation facility with lower rates of death.