The effects of prolonged prone positioning on response and prognosis in patients with acute respiratory distress syndrome: a retrospective cohort study

Table 2 Association between prone duration and clinical outcomes in critically ill patients with ARDS before and after PSM

	Overall	SPP group	PPP group	P	HR/OR(95% CI)
Before PSM	n = 234	n = 133	n = 101
Response rate^a	154 (65.8)	81 (60.9)	73 (72.3)	0.02	1.67 (1.16–2.95)
Length of MV (days)^b	6.26 [2.77, 13.42]	6.40 [2.56, 12.62]	6.17 [3.24, 15.00]	0.64	−0.12 (−0.99 to 1.57)
Length of ICU stay (days)^b	13.00 [7.08, 22.00]	14.00 [8.00, 25.00]	14.00 [6.00, 21.00]	0.56	−1.33 (−5.83 to 3.16)
Length of hospital stay (days)^b	19.00 [10.00, 24.00]	20.00 [11.00, 25.00]	19.00 [9.00, 24.00]	0.41	−1.03 (−3.53 to 1.47)
Re-intubation in ICU, n(%)^c	64 (29.8)	36 (30.5)	28 (28.9)	0.99	0.95 (0.02–1.96)
ICU mortality, n(%)^d	102 (43.6)	60 (45.1)	42 (41.6)	0.06	0.88 (0.42–1.14)
28-day mortality^d	113 (48.3)	67 (50.4)	46 (45.5)	0.04	0.74 (0.57–0.92)
After PSM	n = 162	n = 81	n = 81
Response rate^a	106 (65.4)	49 (60.5)	57 (70.4)	0.03	1.46 (1.23–1.89)
Length of MV (days)^b	7.07 [2.85, 13.68]	7.15 [2.99, 14.00]	6.89 [2.75, 13.61]	0.79	−0.42 (−2.28 to 2.01)
Length of ICU stay (days)^b	13.50 [7.00, 22.32]	14.00 [7.53, 23.00]	13.00 [7.00, 21.00]	0.39	−1.91 (−7.06 to 3.24)
Length of hospital stay (days)^b	20.00 [10.00, 23.00]	19.00 [10.00, 24.00]	19.00 [9.00, 24.00]	0.78	−0.68 (−3.65 to 2.32)
Re-intubation in ICU, n(%)^c	45 (29.8)	22 (29.7)	23 (29.9)	0.86	1.03 (0.22–2.28)
ICU mortality, n(%)^d	76 (46.9)	39 (48.1)	37 (45.7)	0.07	0.70 (0.41–1.21)
28-day mortality^d	81 (50.0)	43 (53.1)	38 (46.9)	0.03	0.53 (0.32–0.85)

ARDS acute respiratory distress syndrome, PSM propensity score matching, MV mechanical ventilation, ICU intensive care unit, HR hazard ratio, OR odds ratio
^aResponse is defined as an increase in PaO₂/FiO₂ ratio by more than 20 mmHg after prone positioning. Logistic regression was used to assess the effect of prone positioning duration on the response
^bLinear regression was performed to evaluate the correlation between prolonged prone positioning and both ICU length of stay and total hospital stay. HR was calculated using the formula HR = e βi. This section includes only patients who received mechanical ventilation
^cLogistic regression was used to evaluate the relationship between prone positioning duration and the reintubation rate
^dCox regression was used to assess the impact of prone positioning duration on ICU and 28-day mortality outcomes. The adjusted variables included those with a P < 0.05 in univariate analysis, as well as confounders selected based on the judgment of experienced clinicians. These variables included age, APACHE II, SOFA, Murray score, prone positioning duration, response status, days on mechanical ventilation, and the time from ICU admission to the start of prone

ISSN: 2052-0492