Table 4 Summary of modifiable factors
Pharmacological or medical measures (modifiable) | Detail or specification / number of studies | ID_study year_author | Results/ findings with M-A; Heterogeneity | Results / findings without M-A | Synthesized finding / Conclusion for the reported outcome |
|---|---|---|---|---|---|
Hyperglycemia / M-A of 3 studies | 2022_Yang_Zi | OR 1.55, CI 95% (0.47–5.12); p = 0.47; I2 = 80%, p = 0.02 |  | The first analysis revealed significant heterogeneity in the association between hyperglycemia and the outcome | |
Hyperglycemia (subgroup) / M-A of 2 studies | 2022_Yang_Zi | OR 2.95 CI 95% (1.70–5.11), p = 0.0001; I2 = 0% p = 0.82 |  | After excluding the study with the largest sample size, substantial changes in overall estimates were observed, and no heterogeneity was found between studies | |
Serum glucose / M-A of 3 studies | 2017_Annoni | OR 3.33 95% CI (− 6.19, 12.84); I2 = 80% p = 0.007 |  | Serum glucose were not associated with ICUAW, significant statistical heterogeneity was found for serum glucose | |
Hyperglycemia / 1 study | 2018_Yang_2 | Not M-A | OR 2.86 CI 95% (1.301–6.296) p = 0.009 | Results of a multivariate analysis of a single independent study indicate an association with increased odds of developing ICUAW | |
Administration of insulin and the measurement of glycaemia / 2 studies | 2017_Sanchéz-Solana | Not M-A | CIPNM incidence: 10% vs. 45% (control) CIPNM incidence: 31% (insulin treatment) vs. 47% (control) | Two articles describe maintaining glycemic control and/or the use of insulin and its association with CIPNM, and these studies include early mobilization therapy in the analysis. Both studies showed a significant decrease in the rate of CIPNM and time on mechanical ventilation | |
Intensive insulin therapy / 3 studies | 2012_Ydemann | Not M-A | (1) OR of 0.49 p < 0.0001 (2) CIPNM incidence from 50.5% to 38.9% p = 0.02 with IIT (3) Reduced diagnosed CIPNM from 74.4 to 48.7% p < 0.0001 | Report of 3 different studies: (1) Pooled data showed that IIT reduced the risk of developing CIPNM. (2) Another study demonstrated a decrease in CIPNM incidence with IIT. (3) Implementation of IIT in two ICUs also significantly reduced diagnosed CIPNM in long-stay patients | |
Glucose levels | 2006_Hohl | Not M-A | Strict blood glucose control (< 6.1 mmol/L) significantly reduced CIPNM incidence from 49 to 25% | Patients with HBG levels in the ICU had a higher incidence of CIPNM, affecting 60% of patients. However, a blood glucose level > 9.4 mmol/L was a positive predictor of paresis, but had low sensitivity (44%) for ruling out CIPNM | |
Use of Neuromuscular blocking agents (NMBAs) Deep sedation (Ramsay score of 6, RASS score of 0 to – 1) [23,24,25,26,27, 29, 30, 38, 39] | Use of NMBAs / M-A of 30 studies | 2023_Bellaver | OR 2.77 CI 95% (1.98–3.88); I2 = 62%, p < 0.00001 |  | Summarized data stratified to RCTs, observational studies and all studies. The size of the effect indicated increased odds of developing ICU-AW According to the GRADE approach, there is a low level of certainty of the evidence |
Use of NMBAs / M-A of 5 studies | 2022_Yang_Zi | OR 1.43 CI 95% (0.92–2.22); I2 = 0%, p = 0.11 |  | Fixed effects model and the combined effect was not statistically significant reported no significant association between NMBAs and ICUAW | |
NMBAs, deep sedation / M-A of 4 studies | 2020_Wei | RR 1.34 CI 95% (0.98–1.84); I2 = 0%, p = 0.898 |  | The incidence of ICUAW was higher in patients who received NMBA treatment. Infusion of NMBA might increase the risk of ICUAW | |
NMBAs, deep sedation / M-A of 4 studies | 2020_Tarazan | RR 1.16 CI 95%; (0.98–1.37); I2 = 0%, p = 0.08 |  | NMBA infusion increases ICUAW risk; however, the 95% CI includes no difference. Moderate certainty of evidence, with an anticipated absolute effect of 346 per 1000 and a risk difference of 55 per 1000 | |
Use of NMBAs / M-A of 3 studies | 2020_Shao | RR 1.19 IC 95% (0.99–1.44); I2 = 0%, p = 0.07 |  | Three studies involving 691 patients provided data on ICUAW. NMBAs did not increase the occurrence of ICU-AW compared to non-NMBA treatment | |
(subgroup MRC score) M-A of 2 studies | 2020_Shao | MD − 2.24 CI 95% (− 6.24–1.76) p = 0.27; I2 = 84% |  | Two studies included 1345 patients reported the MRC score. No statistically significant difference between the two groups (NMBAs experimental v/s placebo) in terms of the MRC scores | |
Use of NMBA / M-A of 5 studies | 2018_Yang_2 | OR, 2.03 CI 95% (1.22–3.40); I2 = 72.9% p = 0.005 |  | A significant association was demonstrated between NMBA use and ICUAW | |
Use of NMBA / M-A of 3 studies | 2017_Annoni | OR 1.61 CI 95% (0.76–3.40); I2 = 74% p = 0.02 |  | Use of neuromuscular NMBA during ICU stay showed a positive association with ICUAW in 4 of 13 studies, and in M-A of 3 studies, with high heterogeneity | |
Use of NMBAs / M-A of 19 studies | 2016_Price | OR, 1.25 IC 95% (1.06–1.48); I2 = 16% p = 0.26 |  | The pooled analysis showed a significant difference in neuromuscular dysfunction: 51% in exposed patients and 39% in controls, with low heterogeneity. The funnel plot suggests possible reporting bias due to small studies with strong associations | |
Use of NMBAs (subgroup lowest RoB studies / 5 studies) | 2016_Price | OR, 1.31 CI 95% (0.91–1.86); I2 = 48% p = 0.10 |  | To show the pooled effect size of studies with the lowest risk of bias (1 RCT, 4 observational studies). The pooled OR was not statistically significant | |
NMBA and sepsis / M-A of 2 studies | 2016_Price | OR 5.36 CI 95% (1.56–18.46); I2 = 1% |  | The M-A of two studies (139 patients with severe sepsis or septic shock) found 83% event rate in exposed vs. 57% in unexposed group. This subgroup had a significant pooled effect size and odds ratio, with minimal heterogeneity | |
NMBAs and asthma / 2 individual studies | 1998_DeJonghe | Not M-A | Â | Two studies involved patients with asthma and/or vecuronium administration. EMG measurement was not systematic, but one study showed a myopathic pattern, and the other found denervation signs in 50% of patients. Prolonged neuromuscular blockade likely contributed to weakness in 20% of patients in the latter study | |
Treatment with corticosteroids / M-A of 8 studies | 2020_Yang_Zi | OR 1.54 CI 95% (0.77–3.09); I2 = 77% p = 0.23 |  | The use of corticosteroids showing significant heterogeneity. Sensitivity analysis did not substantially change overall estimates and heterogeneity persisted | |
Treatment with corticosteroids / M-A of 4 studies | 2018_Yang_2 | OR 1.92 95% CI (0.95–3.88) p > 0.05; I2 = 87.2% p < 0.001 |  | The effect size analysis reported no significant association between corticosteroids and ICUAW | |
Treatment with corticosteroids / M-A of 18 studies | 2018_Yang_1 | OR 1.84 95% IC (1.26–2.67) p = 0.002; I2 = 67.2% p > 0.001 |  | The use of corticosteroids was significantly associated with increased odds of developing ICUAW. The overall incidence of ICUAW was 43% in the corticosteroid group versus 34% in the control group | |
(subgroup clinical weakness)) M-A of 10 studies | 2018_Yang_1 | OR 2.06 95% CI (1.27–3.33), p = 0.003; I2 = 60.6%, p = 0.013 |  | Incidence ICUAW: 39% in the corticosteroid group and 23% in the control group. Significant association with a random effects model considering the observed heterogeneity | |
(subgroup abnormal EMG) M-A of 10 studies | 2018_Yang_1 | OR 1.65; 95% CI (0.92–2.95) p = 0.093; I2 = 70.6%, p < 0.001 |  | No significant association between corticosteroid use and abnormal electrophysiology (event rate: 46% in both groups) | |
Corticosteroids without MV (subgroup) / M-A of 6 studies | 2018_Yang_1 | OR 1.61 95% CI (0.83–3.13) p = 0.161; I2 = 74.4% p = 0.61 |  | Event rate in the corticosteroid group of 31% versus 26% in the control group. No significant association considering the observed heterogeneity | |
Use of corticosteroids / M-A of 3 studies | 2017_Annoni | OR 2.17 95% CI (1.21–3.91); I2 = 45%, p = 0.16 |  | Use of corticosteroids showed a positive association with ICUAW | |
Corticosteroid treatment / 5 individual studies | 2017_Sanchéz-Solana | Not M-A |  | Corticosteroid treatment and CIPNM shows conflicting findings, with most reporting higher CIPNM incidence, one showing an inverse relationship, but no statistically significant association observed | |
Aminoglycoside use / M-A of 3 studies | 2022_Yang_Zi | OR 2.51 95% CI (1.54–4.08); I2 = 0% p = 0.41 |  | A significant association was demonstrated between use aminoglycoside and ICUAW | |
Aminoglycoside use / M-A of 10 studies | 2020_Yang_3 | OR 2.06; 95% IC (1.33–3.21) p = 0.016; I2 = 55.7% |  | The overall effect sizes of the studies revealed a statistically significant relationship between aminoglycoside use and ICUAW, and not to studies limited to patients with abnormal electrophysiology, statistical heterogeneity was obvious | |
(subgroup abnormal electrophysiology) / M-A of 7 studies | 2020_Yang_3 | OR 1.78; 95% CI (0.94–3.39) p = 0.08; I2 = 58.4%, p = 0.025 |  | Seven studies assessed the relationship between aminoglycoside use and abnormal electrophysiology, revealing an incidence of 44% in the aminoglycoside group compared to 39% in the control group. However, the overall effect size did not demonstrate a significant association | |
(subgroup clinical weakness) / M-A of 3 studies | 2020_Yang_3 | OR 2.74; 95% CI (1.83–4.10) p < 0.01; I2 = 0% p = 0.95 |  | Subgroup and sensitivity analyses indicated a significant association between aminoglycoside use and clinical weakness in specific patient populations. Three studies reported an event rate of 46% in the aminoglycoside group compared to 27% in the control group | |
Aminoglycoside use / M-A of 3 studies | 2018_Yang_2 | OR 2.27; 95% CI (1.07–4.81) p < 0.05; I2 = 69.5% p = 0.038 |  | Effect size analysis indicated a statistically significant association between the use of aminoglycosides with ICUAW | |
Aminoglycoside use and SIRS / 3 individual studies | 2006_Hohl | Not M-A |  | No significant differences were found regarding particular drugs and the onset of CIPNM. One prospective study SIRS and the use of aminoglycosides were significantly related (p = 0.03) | |
M-A of 4 studies | 2022_Yang_Zi | OR 1.59, 95% CI (1.11–2.28) p = 0.01; I2 = 0%, p = 0.60 |  | The combined effect from four studies showed a statistically significant association with good literature consistency | |
M-A of 4 studies | 2018_Yang_2 | OR 0.36 95% CI (0.02–7.05) p > 0.05; I2 = 95.2% p < 0.001 |  | There was no effect of RRT on increasing the incidence of ICUAW | |
Use of norepinephrine (NA) [23] | Days of treatment with NA / M-A of 2 studies | 2018_Yang_2 | OR 1.04; 95% CI (0.99–1.09) p > 0.05; I2 = 34.2% p = 0.218 |  | The overall effect size on the association between days of treatment with NA and ICUAW calculated from 2 studies revealed no significant association |
Treatment with NA / 1 study | 2018_Yang_2 | Not M-A | HR 1.30; 95% CI (1.08–1.57) p < 0.05 | Treatment with norepinephrine was found to be a significant risk for developing ICUAW in single study on multivariable analysis | |
Parenteral nutrition / 1 study | 2018_Yang_2 | Not M-A | OR 5.11 95% CI (1.14–22.88) p = 0.02 | Results of a multivariate analysis of a single independent study indicate an association with increased odds of developing ICUAW | |
Energy and/or protein delivery / 6 individual studies | 2018_Lambell | Not M-A |  | A variety of methods were used to assess skeletal muscle mass or TBP. Participants in included studies experienced differing levels of muscle loss (0–22.5%) during the first 2 weeks of ICU admission. No association between energy and protein delivery and changes in skeletal muscle mass were observed. Limited evidence exists regarding this association |