A total of 120 patients (48 at 3 months, 89 at 6 months, and 70 at 12 months) were enrolled in the study. Of these, 54 were evaluated at least once, 45 at least twice, and 21 at 3, 6, and 12 months (Fig. 2). Demographic and clinical characteristics of patients during their ICU stay are presented in Table 1.
Table 1
Demographic characteristics and ICU variables of patients enrolled in the study.
|
Patients
(N = 120)
|
Sex (Male), N. (%)
|
90 (75%)
|
Age (years), Mean (SD)
|
59.8 (11.0)
|
ICU Admission BMI, Kg/m2, Mean (SD)
|
28.9 (4.41)
|
SAPS II, Mean (SD)
|
32.6 (18.5)
|
Use of pre-ICU admission NIV, N. of patients (%)
|
86 (72%)
|
Mechanical ventilation duration (Days), Mean (SD)
|
9.75 (11.4)
|
Use of iNO, N. of patients (%)
|
7 (6%)
|
Prone positioning, N. of patients (%)
|
53 (44%)
|
ECMO, N. of patients (%)
|
6 (5%)
|
Tracheostomy, N. of patients (%)
|
30 (25%)
|
Steroids in ICU, N. of patients (%)
|
88 (73%)
|
Use of NMBA, N. of patients (%)
|
120 (100%)
|
Renal Replacement Therapy, N. of patients (%)
|
5 (4%)
|
Catecholamine, N. of patients (%)
|
47 (39%)
|
Comorbidity, N. of patients (%)
|
|
0
|
11 (9)
|
1
|
37 (31)
|
2
|
39 (33)
|
3
|
18 (15)
|
4
|
10 (8)
|
ICU LOS (Days), Mean (SD)
|
10 (13.2)
|
H LOS (Days), Mean (SD)
|
34.7 (22.7)
|
List of abbreviations: SD, standard deviation. BMI, Body Mass Index. SAPS II, Simplified Acute Physiology Score. NMBA, Neuromuscular Blocking agents. NIV, Non-Invasive Ventilation. iNO, inhaled Nitric Oxide. ICU, Intensive Care Unit, ECMO, Extracorporeal Membrane Oxygenation. H, Hospital. LOS, Length of Stay.
Missing values [number of patients (percentage)]: SAPS II: 22 (18.6%); Use of pre-ICU admission NIV: 5 (4.2%); Mechanical ventilation duration: 6 (5.1%); Use of iNO:5 (4.2%); Pronation:5 (4.2%); ECMO: 6 (5.1%); Tracheostomy: 9 (7.6%); Steroids: 24 (20.3%); CVVH: 6 (5.1%); Catecholamine: 6 (5.1%); Comorbidities: 5 (4%).
Muscle strength improved over time. MVC improved at 12 months (estimate difference [ED] = 4.83 kg when compared with 3 months, P < 0.001) (Table 2). HGS improved at 6 and 12 months as both absolute values in kilograms and percentage predicted value (ED = 4.85%, P = 0.002 at 6 months; ED = 10.44%, P < 0.001 at 12 months). MRCss was ≥ 48 in all patients. CMAP-TA-S100 improved at 6 (ED 1 mV, P = 0.001) and 12 months (ED 1.33 mV, P = 0.001); PENT improved only at 12 months (ED 1.8 mV, P = 0.001). Severe fatigue was reported by 13% of patients at 3 months, 35% at 6 months, and 20% at 1 year, without significant improvement over time. Mental health impairments were documented in few patients, whereas cognitive impairment was present in a significant proportion of patients (27% at 3 months, 18% at 6 months, and 19% at 12 months).
Table 2
Summary of physical, mental health, and cognitive function variables and neuromuscular electrophysiological measurements at different follow-up times.
|
3 Months
(N = 48)
|
6 Months
(N = 89)
|
6 vs 3 months
Estimate (95% CI)
|
p-value
|
12 Months
(N = 70)
|
12 vs 3 months
Estimate (95% CI)
|
p-value
|
BMI (kg/m2), Mean (SD)
|
27.7 (3.50)
|
27.9 (4.37)
|
0.23
(-0.14–0.59)
|
0.932
|
28.1 (3.71)
|
0.79
(0.40–1.18)
|
0.192
|
MRCss,a Mean (SD)
|
58.8 (2.92)
|
59.6 (0.980)
|
0.80
(3.27 - 0.513)
|
0.326
|
59.8 (0.94)
|
1.00
(0.42–1.58)
|
0.321
|
MVC (Kg), Mean (SD)
|
21.8 (9.81)
|
20.8 (8.29)
|
1.38
(-0.15–2.90)
|
0.078
|
24.9 (9.09)
|
4.83
(2.797–6.86)
|
< 0.001
|
Dominant hand grip strength (kg), Mean (SD)
|
31.1 (10.7)
|
30.2 (10.6)
|
2.09
(0.89–3.30)
|
0.001
|
33.9 (10.0)
|
4.18
(2.89–5.46)
|
< 0.001
|
Dominant hand grip strength (% predicted)b, Mean (SD)
|
78.9 (20.5)
|
78.5 (19.1)
|
4.85
(1.08–8.63)
|
0.007
|
85.9 (18.3)
|
10.44
(6.40–14.48)
|
< 0.001
|
CMAP-TA-S100 (mV), Mean (SD)
|
6.18 (2.22)
|
7.14 (1.86)
|
1.00
(0.59–1.42)
|
< 0.001
|
7.39 (2.10)
|
1.33
(0.89–1.77)
|
< 0.001
|
PENT (mV), Mean (SD)
|
6.70 (5.02)
|
6.95 (4.58)
|
0.80
(-0.01–1.61)
|
0.053
|
6.91 (5.03)
|
1.81
(0.92–2.70)
|
< 0.001
|
Fatigue (Fatigue Severity Score ≥ 36), N. (%)
|
6 (13%)
|
31 (35%)
|
2.36
(-0.01–5.43)
|
0.194
|
14 (20%)
|
0.40
(-0.05–2.79
|
0.762
|
Presence of depression (HADS > 7), N. (%)
|
3 (6%)
|
11 (12%)
|
12.34
(2.79–21.89)
|
0.002
|
6 (9%)
|
1.24
(-6.26–5.83)
|
0.932
|
Presence of anxiety (HADS > 7)
|
3 (6%)
|
10 (11%)
|
2.54
(-2.09–7.18)
|
0.567
|
11 (16%)
|
3.32
( -1.21 - 7.86)
|
0.237
|
Post-Traumatic Stress Disorder (PCL-5 > 32), N.(%)
|
2 (4%)
|
8 (9%)
|
30.81
(30.72–30.90)
|
< 0.001
|
3 (4%)
|
14.66
(14.58–14.74)
|
< 0.001
|
Cognitive impairment (MoCA < 26), N. (%)
|
13 (27%)
|
16 (18%)
|
− 2.18
(-5.03–0.67)
|
0.201
|
13 (19%)
|
-3.48
(-7.07 − 0.11)
|
0.061
|
List of Abbreviations: SD, standard deviation. MVC, Maximal Voluntary Contraction. MRCss, Medical Research Council Sum Score. CMAP, amplitude of Compound Muscle Action Potential. TA, Tibialis Anterior. PENT, PEroneal Nerve Test. HADS, Hospital Anxiety and Depression Scale. MoCA, Montreal Cognitive Assessment. PLC-5, PTSD Checklist for DSM-5.
Missing value (3 months-6 months-12months): MRCss: 1 (2.1%)-16 (18.0%)-2 (3.1%); Dominant hand grip strength: 0 (0%)-5 (5.6%)-1 (1.5%); Six-minute-walk test: 1 (2.1%)-18 (20.2%)-4 (6.2%); CMAP: 1 (2.1%)-4 (4.5%)-4 (6.2%); PENT: 1 (2.1%)-0 (0%)-15 (23.1%).
Depression: 0 (0%)- 24 (27%)- 0 (0%); Anxiety 0 (0%)- 24 (27%)- 0 (0%); MoCA: 0 (0%)-4 (4.5%)-0 (0%).
a None of the patients had an MRCss<55.
b Calculated using established reference values provided by Gilbertson L et al. [28]
c Predicted value for the six-minute walk test was calculated according to Enright PL et al. [29]
Multivariable analysis showed that MVC was independently associated with sex (ED 6.89 kg in males vs females, P < 0.001), age (ED -0.18 kg each year, P = 0.012), mechanical ventilation duration (ED -0.12 kg for each day increase of mechanical ventilation, P = 0.044), and CMAP-TA-S100 (ED 1.54 kg for each millivolt increase in CMAP-TA-S100, P < 0.001). The multivariable analysis confirmed the significant improvement in MVC from 3 to 12 months. There was no interaction between follow-up time and the selected variables (Table 3).
Table 3
Effect estimates and corresponding 95% confidence interval (C.I.) for muscle strength prediction using maximal voluntary contraction (MVC in kg) computed using a linear mixed model (LMM) with random intercept. The variable included in the final model were selected using a backward procedure using BIC criterion.
|
Model MVC(Kg)
|
Coefficient
|
Estimates
|
95% C.I.
|
P
|
6 Months vs 3 Months
|
0.13
|
-2.04–2.30
|
0.905
|
12 Months vs 3 Months
|
3.61
|
1.35–5.88
|
0.002
|
Sex (Male)
|
6.89
|
3.54–10.23
|
< 0.001
|
Age (year)
|
-0.18
|
-0.33 – -0.04
|
0.012
|
Steroids in ICU
|
1.81
|
-3.06–6.68
|
0.463
|
MV Duration (days)
|
-0.12
|
-0.23 – -0.00
|
0.044
|
CMAP-TA-S100
|
1.54
|
0.90–2.18
|
< 0.001
|
Presence of anxiety (HADS Score)
|
-0.10
|
-0.39–0.19
|
0.506
|
Presence of PTSD (PCL-5)
|
0.22
|
-0.15–0.60
|
0.243
|
List of Abbreviation: CI, Confidence Interval. MV, Mechanical Ventilation. CMAP, Compound Muscle Action Potential amplitude. TA, Tibialis Anterior. CI, confidence Interval. |
MVC was strongly associated with HGS (ED 0.51 kg [95% confidence interval: 0.39–0.62] increase in HGS for each 1-kg increase in MVC, P < 0.001), but not with MRCss (odds ratio = 0.02, P = 0.352). CMAP-TA-S100 was strongly associated with PENT (ED 0.25 mV increase in PENT for each millivolt increase in CMAP-TA-S100, P < 0.001). Patients with abnormal PENT (< 5.26 mV) had a mean (SD) CMAP-TA-S100 of 6.57 (2.18) mV, whereas patients with normal PENT had a mean (SD) CMAP-TA-S100 of 7.69 (1.70) mV.