key: cord-0873334-fq9ytwkl
authors: Lucchini, Alberto; Russotto, Vincenzo; Barreca, Nicola; Villa, Marta; Casartelli, Giulia; Marcolin, Yelenia; Zyberi, Barbara; Cavagnuolo, Domenico; Verzella, Giacomo; Rona, Roberto; Fumagalli, Roberto; Foti, Giuseppe
title: Short And Long-Term Complications Due To Standard And Extended Prone Position Cycles In Covid-19 Patients
date: 2021-10-29
journal: Intensive Crit Care Nurs
DOI: 10.1016/j.iccn.2021.103158
sha: 939bd4c874ea4e6b6657a7386bbb0c609606af09
doc_id: 873334
cord_uid: fq9ytwkl

OBJECTIVE: to investigate short and long-term complications due to standard (≤24 hours) and extended (> 24 hours) prone position in COVID-19 patients. METHODS: retrospective cohort study conducted in an Italian general ICU. We enrolled patients on invasive mechanical ventilation and treated with prone positioning. We recorded short term complications from Intensive Care Unit (ICU) data chart and long-term complications from the scheduled follow-up visit, three months after ICU discharge. RESULTS: A total of 96 patients were included in the study. Median time for each prone positioning cycle (302 cycles) was equal to 18 (16 - 32) hours. In 37 (38%) patients at least one cycle of extended pronation was implemented. Patients with at least one pressure sore due to prone position were 38 (40%). Patients with pressure sores showed a statistically significative difference in ICU length of stay, mechanical ventilation days, numbers of prone position cycles, total time spent in prone position and the use of extended prone position, compared to patients without pressure sores. All lesions were of low grade. Cheekbones (18%) and Chin (10%) were the most affected sites. Follow-up visit, scheduled three months after ICU discharge, was possible in 58 patients. All patients were able to have all 12 muscle groups examined by Medical Research Council scale examination. No patient reported sensory loss or presence of neuropathic pain for upper limbs. CONCLUSIONS: Extended prone position is feasible and might reduce the workload on the healthcare workers without significant increase of major prone position related complications.

In patients with Adult Respiratory Distress Syndrome (ARDS), prone position (PP) decreased 28-day and 90-day mortality (Guerin et al., 2020) . Several studies have shown its impact on gas exchange (Gattinoni et al., 2001; Guerin et al., 2018; Taccone et al., 2009 ). It's crucial to underline that the improvement of the pO 2 /FiO 2 mmHg ratio value on its own, does not represent the only parameter to evaluate the benefits of prone position. PP, compared to the supine positioning, markedly reduced the overinflated lung areas while promoting alveolar recruitment (Guerin et al., 2013) . These effects may contribute to prevent ventilator-induced lung injury by homogenizing the distribution of stress and strain within the lung and they may represent the mechanisms by which PP reduced mortality independently from its effect on oxygenation. COVID-19 pandemic dramatically increased ICU admission of patients with moderate to severe ARDS, leading to the greatest healthcare crisis of the modern era. The sudden lack of intensive care units (ICU) beds, during the first and the second COVID waves, forced healthcare systems to convert hospital areas to new COVID-19 ICUs. (Bambi et al., 2020; Lucchini et al., 2020b) . This was accompanied by the recruitment of healthcare workers without previous critical care experience. At a time of uncertainty efficacy of the available therapies, PP represented the single intervention with proven efficacy for mortality reduction of COVID-19 related ARDS (Nasa et al., 2021) . During the pandemic, in Italy, PP has been adopted in 61% of the COVID-19 ICU population (Langer et al., 2021) . Despite PP is a well recognized routine intervention for ARDS patients, with a low incidence of short and long term adverse events, PP sessions may be highly demanding for the whole ICU staff considering the high global workload and limited resources. Prone positioning increases the risk of developing hospitalacquired pressure injury (González-Seguel et al., 2021; Sud et al., 2014) . The main preventive strategies to prevent pressure sores development during prone positioning include skin assessment (before, during and after PP), repositioning to offload pressure points on the face and the body, application of dressings, such as hydrocolloids, transparent film, and silicone, to decrease facial skin breakdown (National Pressure Injury Advisory Panel et al.; 2019) . Many ICUs have updated their protocols for pressure sores prevention in PP patients, reflecting necessary changes related to care for COVID-19 patients (Team et al., 2021b) . Johnson and colleagues (2021) recently reported that having a certified wound and skin care nurse on a multi-professional prone-positioning team could help to reduce odds of pressure injuries developing in patients infected with COVID-19. Development of online learning resources (infographics, learning modules and webinar) and dedicated check-list for PP procedure (Santos et al., 2021) are recently reported to improve and disseminate knowledge about prevention of PP complications (Team et al., 2021a) . Preventive strategies recommended by the guidelines suggest to avoid extended use of prone positioning unless required for management of the individual's medical condition. (National Pressure Injury Advisory Panel et al.; 2019) . However, COVID-19 ARDS management requires prone positioning for extended periods of time (Nasa et al., 2021) . A preliminary report of COVID-19 patients proposed the use of prolonged PP (i.e. 36 hours) to improve oxygenation and to reduce nursing workload (Carsetti et al., 2020) . The investigators reported that prolonged PP was feasible and safe. The strategy of prolonged pronation cycles may theoretically be of benefit during the low resources of the pandemic by reducing the overall number of daily handling while offering this intervention to the highest number of patients as feasible for a longer time-interval. On the other hand, this intervention may theoretically be associated with a higher risk of pressure sores, facial oedema or peripheral nerve injuries. To our knowledge, these short and long-term consequences of extended pronation cycles have never been systematically investigated to date. Given the growing number of centers adopting this strategy of extended pronation cycles, it would be relevant to collect information on its advantages and shortcomings for a risk-benefit balance. We designed a retrospective cohort study in order to assess safety of extended pronation cycles in COVID-19 ARDS patients, with a specific focus on pressure sores and peripheral nerve injuries assessed at a 3-month follow-up. The secondary aim was to investigate the modifications of the PaO 2 /FiO 2 mmHg ratio induced by prone position.

This was a retrospective cohort study conducted between February 2020 and January 2021 in the general ICU of San Gerardo University Hospital, Monza, Italy. The ICU consisted of 10 beds before COVID-19 outbreak. During first wave (February-May 2020) ICU beds increased to 21, while in the second wave (October 20-January 21) the beds increased to 19.

We enrolled all consecutive patients with a diagnosis of COVID-19 pneumonia, under invasive mechanical ventilation and prone position. Indication for PP was placed in patients with moderate to severe ARDS and PaO 2 /FiO 2 ratio <150 mmHg and FiO2>0.6. Every prone position cycle was planned for a minimum of 16 hours according to Guerin et al. (Guerin et al., 2013; Nasa et al., 2021) . However, some patients were maintained prone for more than 24 hours to reduce nursing workload (Rezoagli et al., 2021) . We classified as "Standard pronation" every PP cycle lasting ≤24 hours, and as "extended pronation" every PP cycle > 24 hours. We registered all complications related to PP applications (i.e. displacement of indwelling catheters, facial oedema, pressure sores, vomiting, unplanned extubation, airway obstruction due to bronchial secretion retention with need of unplanned bronchoscopy and haemodynamic instability) (Lucchini et al., 2020a) . The National Pressure Ulcer Advisory Panel (NPUAP) score was used to classify the pressure sores (Edsberg et al., 2016) . The Braden Score was used for predicting pressure ulcer risk.

We also calculated the PaO 2 /FiO 2 ratio, at the following time-points: before pronation (PRE-supine step), one hour after pronation (1h-PP step), at the end of pronation (END-PP step) and one hour after supination (POST-supine step).

Every prone positioning manoeuvre was performed according to our ICU protocol and policy, which is described in detail in Supplementary material_S1. Prone position protocol includes the need to protect bony prominences with prophylactic dressing prior to prone positioning, lubricate the eyes and tape them closed, ensure that the endotracheal tube be secured with tapes (National Pressure Injury Advisory Panel et al.; 2019) . Our ICU policy, from January 2020 avoided implementation of "swimming position". Figure 1 showed a patient in prone position. Supplementary material Figure   S2 showed the pillows used in the study period.

We followed up COVID-19 patient three months after ICU discharge. In order to detect neuromuscular injuries as long term complications of PP, we extracted Follow-Up reports on the assessment of muscle peripheral strength by the Medical Research Council (MRC) scale and handheld dynamometry (Council MR. 1976; Hermans G et al., 2012; Parry et al., 2015) . The MRC Scale for muscle strength is a commonly used scale for assessing muscle strength from Grade 5 (normal)

to Grade 0 (no visible contraction). This score was defined as the sum of MRC scores from six muscles in the upper and lower limbs on both sides, so that the score ranged from 60 (normal) to 0 (quadriplegic). Handgrip strength dynamometry has been proposed as a simple and easy diagnostic method for ICU acquired weakness and was performed on both upper limbs (Bragança et al., 2019; Van Aerde et al.,2020) .

Data were collected as part of the "STORM" study (Spallanzani Institute approval number 84/2020; NCT04424992).

We performed the D'Agostino-Pearson test to assess the normal distribution of variables. Variables with normal distribution were reported as mean and standard deviation (SD) and comparison between their means was performed using the Student's t test. Variables without a normal distribution were reported as median and interquartile range (IQR) and comparison between two groups was performed using the Mann-Whitney U-test. Categorical data are reported as frequencies (%) and their difference tested by Chi-Square's or Fisher's exact test as appropriate. One-way repeated measures analysis of variance (rmANOVA) were used to evaluate the differences at the different time points (PRE-supine step: before pronation, 1 h -PP step: one hour after pronation, END-PP Step: at the end of pronation, POST-supine step: one hour after resupination) of the PO 2 /FiO 2 ratio values and respiratory parameters. A p value <0.05 was considered statistically significant. All data were analysed using the Statistical Social Sciences software, version 22.0, for Macintosh (SPSS Inc., Chicago, IL, USA).

From February 20, 2020 to January 31, 2021, 108 patients were admitted to our ICU with a confirmed diagnosis of COVID-19. A total of 96 patients were included in the study. The median age was 59 years (IQR: 53-66), 22 (23%) were female and the median ICU length of stay was 15 days (7-25). At ICU admission patients presented a median Braden score of 10 (10-11) with a median Body Mass Index of 28 (26-33). Comorbidities included hypertension (52% -n=19), type 2 diabetes mellitus (23%, n=17%) and arteriopathies (10% -n=7). Median pO 2 /FiO 2 before the first PP cycle was 116 (82-148), while the median time in hours, from ICU admission to first PP manoeuvre was equal to 8 (4-45) hours. Median time for each PP performed cycle was equal to 18 (16-32) hours. PP was applied for one cycle in 30 (31%) patients, for two cycles in 21 (22%), for three cycles in 16 (17%), while patients who received three or more PP cycles were 29 (30%). In 37 (38%) patients at least one cycle of extended pronation was implemented. The overall time for patients, spent in PP was equal to 48 (31 -101) hours. In 13 (13%) subjects PP was adopted while the patient was connected to veno-venous Extracorporeal Membrane Oxygenation. 79 (82%) patients survived and were discharged from ICU.

Patients with at least one pressure sore due to PP were 38 (40%). Patients with pressure sores showed a statistically significative difference in ICU length of stay, mechanical ventilation days, numbers of PP cycles, total time spent in PP and the use of extended prone position, compared to patients without pressure ulcers. Chin and cheekbones were the most affected sites, where pressure sores were present in 18% (n=17) and 10% (n=10%) of patients. Regarding the pressure ulcer severity, the most frequent stage of NPUAP score was stage II in 93 out of 105 pressure sores (88%), followed by stage I in 9

pressure sores (8.5%) and stage III in 3 sites (3.5%). No pressure sores with stage IV occurred. The characteristics of the sample, and the difference between patients without and with pressure sore development are presented in Table 1 . Incidence of medical device-related pressure injuries in mucosal tissues, because of prone positioning, such as ulcers of the lips were present in 7 patients (7%) caused by endotracheal tubes, but we did not record pressure sores in ala nasi and nostrils due to nasogastric tubes. The total pronation manoeuvres investigated were 302. Extended pronation was most commonly used during the second COVID-19 wave. The overall complications related to PP manoeuvre were haemodynamic instability in 29 (10%) manoeuvres, prolonged arterial desaturation involved 57 (19%) of investigated cycles and bronchial secretions retention, with need of unplanned bronchoscopy, occurred in 33 (11%) manoeuvres. No accidental removal of vascular access devices occurred during the study period, while we observed 2 (1%) nasogastric tube and one (0.5%) endotracheal tube displacement, with no necessity to device repositioning (no completely extubation was occurred). No statistically significant differences in these severe complications were detected between standard and extended pronation cycles. Details about all investigated prone position cycles and related complications are reported in Table 3 . PO 2 /FiO 2 ratio improved during prone position and after resupination compared to baseline (i.e. before prone position) in overall cycles and in extended versus standard pronation groups, as showed in Figure 2 . Table S3 (see Additional file 3) summarizes all mechanical ventilation and oxygenation data.

Follow Up visit, three months after ICU discharge, by anaesthesiologist and ICU nurse was possible in 58 patients. In the remaining 36, the main reasons for the missed visit were: distance from the hospital (patient centralized during COVID and referred to the local hospital) and refusal to visit. No patient reported problems with pressure sores outcomes at the three-month follow-up visit. All patients were able to have all 12 muscle groups examined by MRC examination. Median overall MRC score at Follow Up visit was 60 (59-60). We did not observe significantly difference in MRC score in patients undergoing standard prone position compared to patients with extended pronation [60 (59-60)] versus 60 (58-60) -p=0.395]. For the following tested districts: shoulder adductors, elbow flexors and wrist extensor, the MRC score was always ≥ 3 in all patients. Foot dorsiflexors test was equal to zero in one patient (left side) and in two patients was equal to one (right side). We observed one patient with grade zero in MRC score in left foot dorsiflexors test. Hip flexors test was grade two in one patient (right side). Statistically significance was reported in MRC grade distribution, between patients with and without extended pronation only for the right Elbow flexors test (p=0.028). Table 4 summarize results from MRC all 12 muscle groups examined. After physical examination, no patient reported sensory loss or presence of neuropathic pain for upper limbs. The median overall maximum handgrip dynamometry was 33.0 (24-37) kg-force. Finally, we did not observe any difference between standard and extended pronation groups in handgrip dynamometry results [33(25.0-37) vs. 29 (20-39) kg-force -p=0.679].

Prone position for ventilated patients with COVID-19 ARDS was strongly suggested by experts, for a duration of 16-24 h per session, similar to the indication in non-COVID-19-related ARDS (Nasa et al., 2021) . In our study the rate of pressure sores was similar to the results of the studies published before the Covid-era (Lucchini et al., 2018a and 2020a; Sud et al., 2014) , but lower if we consider recent studies on PP and COVID-19 patients (Binda et al., 2021; Douglas et al., 2021; Ibarra et al., 2020) . Binda and colleagues, in a retrospective study in a single ICU in Italy, involved 63 COVID-19 patients with a total of 219 proning cycles, report an incidence of prone-related pressure ulcers in about 30% of patients (Binda et al., 2021) . Ibarra and colleagues, in Spain, published a case-series of patients on invasive mechanical ventilation and PP therapy with 77% of the patients presented pressure sores due to PP (Ibarra et al., 2021) . The frequency of pressure sores in the face area, found in our study was lower than those reported in two systematic reviews (Abroug et al., 2008; Sud et al., 2014) and in a narrative review (González-Seguel et al., 2021) in which pressure ulcers occurred in 34%, 43% and 19% of patients, respectively. Important to underline that we didn't detected grade III in the reported pressure sores developed on face area. Low-grade facial pressure sores have less serious consequences, and none of our patients needed special care or treatment. All the pressure sores were managed with dressings, achieving wound healing by secondary intention in all the survivors (Perrillat et al., 2020; Shearer et al., 2021) . Once their condition improved and prone positioning was no longer performed, the skin fully recovered in all patients. No patient reported any problems with pressure sore outcomes at the follow-up visit, after ICU discharge.

Comparing the incidence of pressure sores with two studies published by our group (Lucchini et al 2018a and 2020a) in the last 5 years, however, we must note that the global incidence has slightly increased. In our previously study on PP complications, ICU team was composed only by ICU nurses with long experience in nursing and management of ARDS patients. During the COVID-era, human and material health care resources had to be adapted to an unpredictable new "war" scenario in a very short period of time (Bambi et al., 2020; Lucchini et al., 2020b and 2020c) . In a few days, the growing demand for COVID-19 patients ICU beds in our region, forced us, during the first wave (February-April 20) to realise 11 new beds in the operating theatre in addition to our 10 bed ECMO unit. The original ICU staff were divided between the new and the old ICU beds. 33 nurses from the operating theatre were also recruited as new ICU staff. In this context, it is understandable that the complications of this procedure could increase. In order to guarantee a safe PP procedure, at least three experienced operators (critical care nurses and/or Intensivist physicians) were present during each prone position manoeuvre. During second wave ICU only 10 nurses from the operating theatre were recruited. Even for this shortage of nurses, the use of extended prone position became frequent in our ICU, during the second pandemic wave (Rezoagli et al., 2021) . In addition to the reduced nurses' experience, the nurse-to-patient ratio changed between the first and second COVID-19 waves.

In the first wave there was always, a nurse-to patient ratio equal to 1: 2. In the second wave, the nurseto-patient ratio of 1:2 was always guaranteed only for ECMO patients. In the other patients, the nurseto-patient ratio was 1:2,5.

Patients undergoing "extended prone position" in our study, trended to a higher number of proning cycles [3 (2-4) versus 2 (1-4) -p=0.017] and to a longer time spent in prone position during their ICU stay compared to standard PP group [33 (18-64) versus 85 (43-136), p=0.0001]. Oxygenation improvement in our sample seemed to be higher during "extended pronation" than during standard pronation, and after resupination compared to baseline. The risk of developing a higher incidence of pressure sores in the prolonged extended pronation is an issue that should be cautiously monitored in present and future studies (National Pressure Injury Advisory Panel et al.; 2019) . In our sample we observed an increased rate of pressure sores (51% extended PP vs. 32% standard PP -p=0.0001), but we did not observe an increase of the others PP complications.

In previous published metanalyses and narrative reviews (Abroug et al., 2008; González-Seguel et al.,2021; Sud et al., 2014; ) the total percentage of airway related complications described for prone position were present in a range between 20 and 40% of investigated cycles. We did not observe any unplanned extubation. In all patients, our ICU policy provided that the artificial airway was secured with a 5 centimeters canvas tape placed upon a thin hydrocolloid (Lucchini et al., 2018b) . Before every pronation, the tape was replaced in order to guarantee better stability (National Pressure Injury Advisory Panel et al.; 2019) . Moreover, before pronation, the tube was displaced on the side of the mouth not leaning on the pillow (for example when the head was rotated on the right side, the tube was fixed on the left one, and vice versa). Our protocol included the use of thin hydrocolloid as protective skin coverings, under devices and over bony prominences (forehead, cheeks, chin, iliac crest, ribs and patella and tibial plateau) (Lucchini et al., 2020a; Peko et al., 2020) . The regular use of hydrocolloids, may be a beneficial factor in decreasing skin breakdown in PP (National Pressure Injury Advisory Panel et al.; 2019) . The current evidence suggests that positioning devices to offload pressure points on the face and body are a useful addition within an overall PU prevention strategy (Binda et al., 2021; Peko et al., 2021; Rodríguez-Huerta et al., 2021) .

Finally, no long term adverse events due to PP have been observed during the follow-up visit. Nerve injuries are uncommon following prone positioning (Goettler et al., 2002) , but brachial plexus injuries have been recently reported (Brugliera et al., 2021; Douglas et al., 2021; Miller et al., 2021) .

Brachial plexus injuries could occur in PP patients when shoulders are positioned in abduction with external rotation and posteriorly displaced and this position causes compression and stretching of the brachial plexus, usually affecting the upper nerve roots (Bozentka et al., 1998) . Recently Miller et al. (2021) , in a study involving 256 COVID-19 patients with PP, reported that in 12 (5%) patients the ulnar nerve was injured, and in 11 patients, an injury was presents at the cords of the brachial plexus. 30 (10%) patients reported neuropathic pain, and all patients presented with motor weakness. Brugliera et al, (2021) reported 7 (5%) cases of brachial plexopathy out of 135 patients who had undergone in PP, in an Italian ICU. Douglas et al. (2021) reported brachial plexus palsies in five (8.2%) of 61 patients treated with extended pronation. In all these studies, the ICU teams used a standard protocol for PP, with implementation of the "swimming position", always combined with alternating arm reposition, performed between 2 or 4 hours. Generally, authors that reported the "swimming position" adoption, are unable to extract exactly data on the frequency of arm cycling while patients were proned. The "swimming position" involves raising one arm on the same side to which the head is facing while placing the other arm by the patient's side. The shoulder should be abducted to 80 degrees and the elbow flexed 90 degrees on the raised arm. This is, however, a potential stretch position for the ulnar nerve at the elbow (Bozentka et al., 1998) . Before the study period, the prone position protocol in our unit involved the use of the "swimming position". In this position, our protocol before December 2019, required that only the left arm was raised (Lucchini et al., 2018a and 2020a) . In an internal audit (December 2019) we reported two cases of brachial plexus injury detected at the scheduled follow-up visit. For this reason, in order to investigate the possible rule of "swimming position" in those adverse events developments, we decided, starting from January 2020, avoid it during PP. Shortly after this decision, the COVID-era began in Italy and in our ICU (Giani et al., 2020; Lucchini et al., 2020b) . All enrolled patients in our study were pronated by 

Several limitations should be considered in the interpretation and generalization of these findings.

First, one limitation may be considered the single-centre, retrospective design. Second, data on complications were recorded in a pandemic period characterized by a reduction in the usual nursepatient ratio, and an increased workload with a potential for underestimation of pressure sores detection. However, this is the first study which primarily investigated and reported the most expected complications from prolonged pronation cycles. Third, follow-up visit was possible only in 58 patients and we cannot exclude the presence of complications in patients missed to follow-up.

This retrospective study has shown that, even during a pandemic surge, coupled with limited resources, prone position could be applied with an acceptable increase in complications rate.

Extended prone positioning is feasible although with a slight increase in the incidence of low severity pressure sores respect to standard prone position and no major serious complications have been 

Buttocks -median NPUAP grade 2 (1-2) 2 (1-2) --- Table 1 Title : Descriptive characteristics of enrolled patients. with and without pressure sores Table 3 Title : Description of prone positioning cycles and complications Left 0 (0%) 0 (0%) 0 (0%) 1 (2%) 12 (21%) 45 (78%) Shoulder adductors Right 0 (0%) 0 (0%) 0 (0%) 2 (3%) 8 (14%) 48 (83%) Left 0 (0%) 0 (0%) 0 (0%) 0 (0%) 8 (14%) 50 (86%) Elbow flexors Right 0 (0%) 0 (0%) 0 (0%) 1 (2%) 6 (10%) 51 (88%) Left 0 (0%) 0 (0%) 0 (0%) 0 (0%) 7 (12%) 51 (88%) Wrist extensor Right 0 (0%) 0 (0%) 0 (0%) 1 (2%) 6 (10%) 51 (88%) Left 0 (0%) 0 (0%) 0 (0%) 0 (0%) 9 (16%) 49 (84%) Hip flexors Right 0 (0%) 0 (0%) 1 (2%) 0 (0%) 8 (14%) 49 (84%) Left 0 (0%) 0 (0%) 0 (0%) 0 (0%) 9 (16%) 49 (84%) Knee extensors Right 0 (0%) 0 (0%) 0 (0%) 0 (0%) 8 (14%) 50 (86%) Left 1 (2%) 0 (0%) 0 (0%) 1 (2%) 9 (16%) 47 (81%) Foot dorsiflexors Right 0 (0%) 2 (3%) 0 (0%) 1 (2%) 8 (14%) 47 (81%) 

Median number of body areas with Pressure sores related to PP 1 (1 -2) ---1 (1-2) ---Forehead 4 (4%) ---4 (10%) p<0.0001

Cheekbones 10 (10%) ---10 (26%) p<0.0001

Chin 17 (18%) ---17 (44%) p<0.0001

Lips 7 (7%) ---7 (18%) 0.001

Thorax 9 (9%) ---9 (24%) p<0.0001

Abdomen 1 (1%) ---1 (3%) 0.214

Iliac crests 3 (3%) ---3 (8%) 0.03

Tibial plateau 3 (3%) ---3 (8%) 0.03

Knees 9 (9%) ---9 (24%) p<0.0001

Other sites with Pressure sores 4 (4%) ---4 (11%) 0.012

Occiput 1 (1%) ---1 0.214

Shoulder blades 14 (15%) 11 (19%) 3 (8%) 0.133

Sacrum 23 (24%) 9 (16%) 14 (37%) 0.017

Buttocks 2 (2%) 2 (3%) ---0.247

Buttocks -median NPUAP grade 2 (1-2) 2 (1-2) --- Table 1 Title : Descriptive characteristics of enrolled patients. with and without pressure sores Table 3 Title : Description of prone positioning cycles and complications Left 0 (0%) 0 (0%) 0 (0%) 1 (2%) 12 (21%) 45 (78%) Shoulder adductors Right 0 (0%) 0 (0%) 0 (0%) 2 (3%) 8 (14%) 48 (83%) Left 0 (0%) 0 (0%) 0 (0%) 0 (0%) 8 (14%) 50 (86%) Elbow flexors Right 0 (0%) 0 (0%) 0 (0%) 1 (2%) 6 (10%) 51 (88%) Left 0 (0%) 0 (0%) 0 (0%) 0 (0%) 7 (12%) 51 (88%) Wrist extensor Right 0 (0%) 0 (0%) 0 (0%) 1 (2%) 6 (10%) 51 (88%) Left 0 (0%) 0 (0%) 0 (0%) 0 (0%) 9 (16%) 49 (84%) Hip flexors Right 0 (0%) 0 (0%) 1 (2%) 0 (0%) 8 (14%) 49 (84%) Left 0 (0%) 0 (0%) 0 (0%) 0 (0%) 9 (16%) 49 (84%) Knee extensors Right 0 (0%) 0 (0%) 0 (0%) 0 (0%) 8 (14%) 50 (86%) Left 1 (2%) 0 (0%) 0 (0%) 1 (2%) 9 (16%) 47 (81%) et al., 2020) . Several studies have shown its impact on gas exchange (Gattinoni et al., 2001; Guerin et al., 2018; Taccone et al., 2009 ). It's crucial to underline that the improvement of the pO 2 /FiO 2 mmHg ratio value on its own, does not represent the only parameter to evaluate the benefits of prone position. PP, compared to the supine positioning, markedly reduced the overinflated lung areas while promoting alveolar recruitment (Guerin et al., 2013) . These effects may contribute to prevent ventilator-induced lung injury by homogenizing the distribution of stress and strain within the lung and they may represent the mechanisms by which PP reduced mortality independently from its effect on oxygenation. COVID-19 pandemic dramatically increased ICU admission of patients with moderate to severe ARDS, leading to the greatest healthcare crisis of the modern era. The sudden lack of intensive care units (ICU) beds, during the first and the second COVID waves, forced healthcare systems to convert hospital areas to new COVID-19 ICUs. (Bambi et al., 2020; Lucchini et al., 2020b) . This was accompanied by the recruitment of healthcare workers without previous critical care experience. At a time of uncertainty efficacy of the available therapies, PP represented the single intervention with proven efficacy for mortality reduction of COVID-19 related ARDS (Nasa et al., 2021) . During the pandemic, in Italy, PP has been adopted in 61% of the COVID-19 ICU population (Langer et al., 2021) . Despite PP is a well recognized routine intervention for ARDS patients, with a low incidence of short and long term adverse events, PP sessions may be highly demanding for the whole ICU staff considering the high global workload and limited resources. Prone positioning increases the risk of developing hospitalacquired pressure injury (González-Seguel et al., 2021; Sud et al., 2014) . The main preventive strategies to prevent pressure sores development during prone positioning include skin assessment (before, during and after PP), repositioning to offload pressure points on the face and the body, application of dressings, such as hydrocolloids, transparent film, and silicone, to decrease facial skin breakdown (National Pressure Injury Advisory Panel et al.; 2019) . Many ICUs have updated their protocols for pressure sores prevention in PP patients, reflecting necessary changes related to care for COVID-19 patients (Team et al., 2021b) . Johnson and colleagues (2021) recently reported that having a certified wound and skin care nurse on a multi-professional prone-positioning team could help to reduce odds of pressure injuries developing in patients infected with COVID-19. Development of online learning resources (infographics, learning modules and webinar) and dedicated check-list for PP procedure (Santos et al., 2021) are recently reported to improve and disseminate knowledge about prevention of PP complications (Team et al., 2021a) . Preventive strategies recommended by the guidelines suggest to avoid extended use of prone positioning unless required for management of the individual's medical condition. (National Pressure Injury Advisory Panel et al.; 2019) . However, COVID-19 ARDS management requires prone positioning for extended periods of time (Nasa et al., 2021) . A preliminary report of COVID-19 patients proposed the use of prolonged PP (i.e. 36 hours)

to improve oxygenation and to reduce nursing workload (Carsetti et al., 2020) . The investigators reported that prolonged PP was feasible and safe. The strategy of prolonged pronation cycles may theoretically be of benefit during the low resources of the pandemic by reducing the overall number of daily handling while offering this intervention to the highest number of patients as feasible for a longer time-interval. On the other hand, this intervention may theoretically be associated with a higher risk of pressure sores, facial oedema or peripheral nerve injuries. To our knowledge, these short and long-term consequences of extended pronation cycles have never been systematically investigated to date. Given the growing number of centers adopting this strategy of extended pronation cycles, it would be relevant to collect information on its advantages and shortcomings for a risk-benefit balance. We designed a retrospective cohort study in order to assess safety of extended pronation cycles in COVID-19 ARDS patients, with a specific focus on pressure sores and peripheral nerve injuries assessed at a 3-month follow-up. The secondary aim was to investigate the modifications of the PaO 2 /FiO 2 mmHg ratio induced by prone position.

This was a retrospective cohort study conducted between February 2020 and January 2021 in the general ICU of San Gerardo University Hospital, Monza, Italy. The ICU consisted of 10 beds before COVID-19 outbreak. During first wave (February-May 2020) ICU beds increased to 21, while in the second wave (October 20-January 21) the beds increased to 19.

We enrolled all consecutive patients with a diagnosis of COVID-19 pneumonia, under invasive mechanical ventilation and prone position. Indication for PP was placed in patients with moderate to severe ARDS and PaO 2 /FiO 2 ratio <150 mmHg and FiO2>0.6. Every prone position cycle was planned for a minimum of 16 hours according to Guerin et al. (Guerin et al., 2013; Nasa et al., 2021) . However, some patients were maintained prone for more than 24 hours to reduce nursing workload (Rezoagli et al., 2021) . We classified as "Standard pronation" every PP cycle lasting ≤24 hours, and as "extended pronation" every PP cycle > 24 hours. We registered all complications related to PP applications (i.e. displacement of indwelling catheters, facial oedema, pressure sores, vomiting, unplanned extubation, airway obstruction due to bronchial secretion retention with need of unplanned bronchoscopy and haemodynamic instability) (Lucchini et al., 2020a) . The National Pressure Ulcer Advisory Panel (NPUAP) score was used to classify the pressure sores (Edsberg et al., 2016) . The Braden Score was used for predicting pressure ulcer risk.

We also calculated the PaO 2 /FiO 2 ratio, at the following time-points: before pronation (PRE-supine step), one hour after pronation (1h-PP step), at the end of pronation (END-PP step) and one hour after supination (POST-supine step).

Every prone positioning manoeuvre was performed according to our ICU protocol and policy, which is described in detail in Supplementary material_S1. Prone position protocol includes the need to protect bony prominences with prophylactic dressing prior to prone positioning, lubricate the eyes and tape them closed, ensure that the endotracheal tube be secured with tapes (National Pressure Injury Advisory Panel et al.; 2019) . Our ICU policy, from January 2020 avoided implementation of "swimming position". Figure 1 showed a patient in prone position. Supplementary material Figure   S2 showed the pillows used in the study period.

We followed up COVID-19 patient three months after ICU discharge. In order to detect neuromuscular injuries as long term complications of PP, we extracted Follow-Up reports on the assessment of muscle peripheral strength by the Medical Research Council (MRC) scale and handheld dynamometry (Council MR. 1976; Hermans G et al., 2012; Parry et al., 2015) . The MRC Scale for muscle strength is a commonly used scale for assessing muscle strength from Grade 5 (normal)

to Grade 0 (no visible contraction). This score was defined as the sum of MRC scores from six muscles in the upper and lower limbs on both sides, so that the score ranged from 60 (normal) to 0 (quadriplegic). Handgrip strength dynamometry has been proposed as a simple and easy diagnostic method for ICU acquired weakness and was performed on both upper limbs (Bragança et al., 2019; Van Aerde et al.,2020) .

Data were collected as part of the "STORM" study (Spallanzani Institute approval number 84/2020; NCT04424992).

We performed the D'Agostino-Pearson test to assess the normal distribution of variables. Variables with normal distribution were reported as mean and standard deviation (SD) and comparison between their means was performed using the Student's t test. Variables without a normal distribution were reported as median and interquartile range (IQR) and comparison between two groups was performed using the Mann-Whitney U-test. Categorical data are reported as frequencies (%) and their difference tested by Chi-Square's or Fisher's exact test as appropriate. One-way repeated measures analysis of variance (rmANOVA) were used to evaluate the differences at the different time points (PRE-supine step: before pronation, 1 h -PP step: one hour after pronation, END-PP Step: at the end of pronation, POST-supine step: one hour after resupination) of the PO 2 /FiO 2 ratio values and respiratory parameters. A p value <0.05 was considered statistically significant. All data were analysed using the Statistical Social Sciences software, version 22.0, for Macintosh (SPSS Inc., Chicago, IL, USA). 

Patients with at least one pressure sore due to PP were 38 (40%). Patients with pressure sores showed a statistically significative difference in ICU length of stay, mechanical ventilation days, numbers of PP cycles, total time spent in PP and the use of extended prone position, compared to patients without pressure ulcers. Chin and cheekbones were the most affected sites, where pressure sores were present in 18% (n=17) and 10% (n=10%) of patients. Regarding the pressure ulcer severity, the most frequent stage of NPUAP score was stage II in 93 out of 105 pressure sores (88%), followed by stage I in 9

pressure sores (8.5%) and stage III in 3 sites (3.5%). No pressure sores with stage IV occurred. The characteristics of the sample, and the difference between patients without and with pressure sore development are presented in Table 1 . Incidence of medical device-related pressure injuries in mucosal tissues, because of prone positioning, such as ulcers of the lips were present in 7 patients (7%) caused by endotracheal tubes, but we did not record pressure sores in ala nasi and nostrils due to nasogastric tubes. Table 3 . PO 2 /FiO 2 ratio improved during prone position and after resupination compared to baseline (i.e. before prone position) in overall cycles and in extended versus standard pronation groups, as showed in Figure 2 . 

Prone position for ventilated patients with COVID-19 ARDS was strongly suggested by experts, for a duration of 16-24 h per session, similar to the indication in non-COVID-19-related ARDS (Nasa et al., 2021) . In our study the rate of pressure sores was similar to the results of the studies published before the Covid-era (Lucchini et al., 2018a and 2020a; Sud et al., 2014) , but lower if we consider recent studies on PP and COVID-19 patients (Binda et al., 2021; Douglas et al., 2021; Ibarra et al., 2020) . Binda and colleagues, in a retrospective study in a single ICU in Italy, involved 63 COVID-19 patients with a total of 219 proning cycles, report an incidence of prone-related pressure ulcers in about 30% of patients (Binda et al., 2021) . Ibarra and colleagues, in Spain, published a case-series of patients on invasive mechanical ventilation and PP therapy with 77% of the patients presented pressure sores due to PP (Ibarra et al., 2021) . The frequency of pressure sores in the face area, found in our study was lower than those reported in two systematic reviews (Abroug et al., 2008; Sud et al., 2014) and in a narrative review (González-Seguel et al., 2021) in which pressure ulcers occurred in 34%, 43% and 19% of patients, respectively. Important to underline that we didn't detected grade III in the reported pressure sores developed on face area. Low-grade facial pressure sores have less serious consequences, and none of our patients needed special care or treatment. All the pressure sores were managed with dressings, achieving wound healing by secondary intention in all the survivors (Perrillat et al., 2020; Shearer et al., 2021) . Once their condition improved and prone positioning was no longer performed, the skin fully recovered in all patients. No patient reported any problems with pressure sore outcomes at the follow-up visit, after ICU discharge.

Comparing the incidence of pressure sores with two studies published by our group (Lucchini et al 2018a and 2020a) in the last 5 years, however, we must note that the global incidence has slightly increased. In our previously study on PP complications, ICU team was composed only by ICU nurses with long experience in nursing and management of ARDS patients. During the COVID-era, human and material health care resources had to be adapted to an unpredictable new "war" scenario in a very short period of time (Bambi et al., 2020; Lucchini et al., 2020b and 2020c) . In a few days, the growing demand for COVID-19 patients ICU beds in our region, forced us, during the first wave (February-April 20) to realise 11 new beds in the operating theatre in addition to our 10 bed ECMO unit. The original ICU staff were divided between the new and the old ICU beds. 33 nurses from the operating theatre were also recruited as new ICU staff. In this context, it is understandable that the complications of this procedure could increase. In order to guarantee a safe PP procedure, at least three experienced operators (critical care nurses and/or Intensivist physicians) were present during each prone position manoeuvre. During second wave ICU only 10 nurses from the operating theatre were recruited. Even for this shortage of nurses, the use of extended prone position became frequent in our ICU, during the second pandemic wave (Rezoagli et al., 2021) . In addition to the reduced nurses' experience, the nurse-to-patient ratio changed between the first and second COVID-19 waves.

In the first wave there was always, a nurse-to patient ratio equal to 1: 2. In the second wave, the nurseto-patient ratio of 1:2 was always guaranteed only for ECMO patients. In the other patients, the nurseto-patient ratio was 1:2,5.

Patients undergoing "extended prone position" in our study, trended to a higher number of proning Oxygenation improvement in our sample seemed to be higher during "extended pronation" than during standard pronation, and after resupination compared to baseline. The risk of developing a higher incidence of pressure sores in the prolonged extended pronation is an issue that should be cautiously monitored in present and future studies (National Pressure Injury Advisory Panel et al.; 2019) . In our sample we observed an increased rate of pressure sores (51% extended PP vs. 32% standard PP -p=0.0001), but we did not observe an increase of the others PP complications.

In previous published metanalyses and narrative reviews (Abroug et al., 2008; González-Seguel et al.,2021; Sud et al., 2014; ) the total percentage of airway related complications described for prone position were present in a range between 20 and 40% of investigated cycles. We did not observe any unplanned extubation. In all patients, our ICU policy provided that the artificial airway was secured with a 5 centimeters canvas tape placed upon a thin hydrocolloid (Lucchini et al., 2018b) . Before every pronation, the tape was replaced in order to guarantee better stability (National Pressure Injury

Advisory Panel et al.; 2019) . Moreover, before pronation, the tube was displaced on the side of the mouth not leaning on the pillow (for example when the head was rotated on the right side, the tube was fixed on the left one, and vice versa). Our protocol included the use of thin hydrocolloid as protective skin coverings, under devices and over bony prominences (forehead, cheeks, chin, iliac crest, ribs and patella and tibial plateau) (Lucchini et al., 2020a; Peko et al., 2020) . The regular use of hydrocolloids, may be a beneficial factor in decreasing skin breakdown in PP (National Pressure Injury Advisory Panel et al.; 2019) . The current evidence suggests that positioning devices to offload pressure points on the face and body are a useful addition within an overall PU prevention strategy (Binda et al., 2021; Peko et al., 2021; Rodríguez-Huerta et al., 2021) .

Finally, no long term adverse events due to PP have been observed during the follow-up visit. Nerve injuries are uncommon following prone positioning (Goettler et al., 2002) , but brachial plexus injuries have been recently reported (Brugliera et al., 2021; Douglas et al., 2021; Miller et al., 2021) .

Brachial plexus injuries could occur in PP patients when shoulders are positioned in abduction with external rotation and posteriorly displaced and this position causes compression and stretching of the brachial plexus, usually affecting the upper nerve roots (Bozentka et al., 1998) . Recently Miller et al. (2021) , in a study involving 256 COVID-19 patients with PP, reported that in 12 (5%) patients the ulnar nerve was injured, and in 11 patients, an injury was presents at the cords of the brachial plexus.

30 (10%) patients reported neuropathic pain, and all patients presented with motor weakness. Brugliera et al, (2021) reported 7 (5%) cases of brachial plexopathy out of 135 patients who had undergone in PP, in an Italian ICU. Douglas et al. (2021) reported brachial plexus palsies in five (8.2%) of 61 patients treated with extended pronation. In all these studies, the ICU teams used a standard protocol for PP, with implementation of the "swimming position", always combined with alternating arm reposition, performed between 2 or 4 hours. Generally, authors that reported the "swimming position" adoption, are unable to extract exactly data on the frequency of arm cycling while patients were proned. The "swimming position" involves raising one arm on the same side to which the head is facing while placing the other arm by the patient's side. The shoulder should be abducted to 80 degrees and the elbow flexed 90 degrees on the raised arm. This is, however, a potential stretch position for the ulnar nerve at the elbow (Bozentka et al., 1998) . Before the study period, the prone position protocol in our unit involved the use of the "swimming position". In this position, our protocol before December 2019, required that only the left arm was raised (Lucchini et al., 2018a and 2020a) . In an internal audit (December 2019) we reported two cases of brachial plexus injury detected at the scheduled follow-up visit. For this reason, in order to investigate the possible rule of "swimming position" in those adverse events developments, we decided, starting from January 2020, avoid it during PP. Shortly after this decision, the COVID-era began in Italy and in our ICU (Giani et al., 2020; Lucchini et al., 2020b) . All enrolled patients in our study were pronated by 

Several limitations should be considered in the interpretation and generalization of these findings.

First, one limitation may be considered the single-centre, retrospective design. Second, data on complications were recorded in a pandemic period characterized by a reduction in the usual nursepatient ratio, and an increased workload with a potential for underestimation of pressure sores detection. However, this is the first study which primarily investigated and reported the most expected complications from prolonged pronation cycles. Third, follow-up visit was possible only in 58 patients and we cannot exclude the presence of complications in patients missed to follow-up.

This retrospective study has shown that, even during a pandemic surge, coupled with limited resources, prone position could be applied with an acceptable increase in complications rate. 

The effect of prone positioning in acute respiratory distress syndrome or acute lung injury: a meta-analysis: areas of uncertainty and recommendations for research

New Issues in Nursing Management During the COVID-19 Pandemic in Italy

S2e guideline: positioning and early mobilisation in prophylaxis or therapy of pulmonary disorders : Revision 2015: S2e guideline of the German Society of Anaesthesiology and Intensive Care Medicine (DGAI)

Complications of prone positioning in patients with COVID-19: A cross-sectional study. Intensive Crit Care Nurs

Cubital Tunnel Syndrome Pathophysiology: Clinical Orthopaedics and Related Research®

Use of handgrip dynamometry for diagnosis and prognosis assessment of intensive care unit acquired weakness: A prospective study

Nerve Compression Injuries After Prolonged Prone Position Ventilation in Patients With SARS-CoV-2: A Case Series

Nursing issues in enteral nutrition during prone position in critically ill patients: A systematic review of the literature

Prolonged prone position ventilation for SARS-CoV-2 patients is feasible and effective

Effects of thoraco-pelvic supports during prone position in patients with acutelung injury/acute respiratory distress syndrome: a physiological study

Aids to the Investigation of Peripheral Nerve Injuries. London: Her Majesty's Stationary Office

Safety and outcomes of prolonged usual care prone position mechanical ventilation to treat acute Coronavirus Disease 2019 hypoxemic respiratory failure

Revised national pressure ulcer advisory panel pressure injury staging system

Effect of prone positioning on the survival of patients with acuterespiratory failure

Exuberant Plasmocytosis in Bronchoalveolar Lavage Specimen of the First Patient Requiring Extracorporeal Membrane Oxygenation for SARS-CoV-2 in Europe

Adverse Events of Prone Positioning in Mechanically Ventilated Adults with Acute Respiratory Distress Syndrome. Respir Care

Brachial plexopathy after prone positioning

Critical care utilization for the COVID-19 outbreak in Lombardy, Italy: early experience and forecast during an emergency response

Prone positioning in severe acute respiratory distress syndrome

A prospective international observational prevalence study on pronepositioning of ARDS patients: the APRONET (ARDS Prone Position Network)study

Prone position in ARDS patients: why, when, how and for whom

Management of critically ill patients with COVID-19: suggestions and instructions from the coordination of intensive care units of Lombardy

Interobserver agreement of Medical Research Council sum-score and handgrip strength in the intensive care unit

Pressure Injury Outcomes of a Prone-Positioning Protocol in Patients With COVID and ARDS

Prone position pressure sores in the COVID-19 pandemic: The Madrid experience

Prone position in intubated, mechanically ventilated patients with COVID-19: a multi-centric study of more than 1000 patients

Enteral nutrition during prone positioning in mechanically ventilated patients

Application of prone position in hypoxaemic patients supported by veno-venous ECMO. Intensive Crit Care Nurs

Unplanned extubations in general intensive care unit: A nine-year retrospective analysis

Prone position in acute respiratory distress syndrome patients: a retrospective analysis of complications

Nursing Activities Score is increased in COVID-19 patients. Intensive Crit Care Nurs

Nursing workload in the COVID-19 era. Intensive Crit Care Nurs

Brachial Plexus Neuropathies During the COVID-19 Pandemic: A Retrospective Case Series of 15 Patients in Critical Care

Expert consensus statements for the management of COVID-19-related acute respiratory failure using a Delphi method

European Pressure Ulcer Advisory Panel, and Pan Pacific Pressure Injury Alliance. Prevention and treatment of pressure ulcers/injuries: Clinical practice guideline

A new two-tier strength assessmen tapproach to the diagnosis of weakness in intensive care: an observational study

Protecting prone positioned patients from facial pressure ulcers using prophylactic dressings: a timely biomechanical analysis in the context of the COVID-19 pandemic

Facial pressure ulcers in COVID-19 patients undergoing prone positioning: how to prevent an underestimated epidemic?

Difference between prolonged versus standard duration of prone position in COVID-19 patients: a retrospective study

Nursing care and prevalence of adverse events in prone position: characteristics of mechanically ventilated patients with severe SARS-CoV-2 pulmonary infection

COVID-19 patients in prone position: validation of instructional materials for pressure injury prevention

Facial pressure injuries from prone positioning in the COVID-19 era

Effect of prone positioning during mechanical ventilation on mortality among patients with acute respiratory distress syndrome: a systematic review and meta-analysis

Prone positioning in patients with moderate and severe acute respiratory distress syndrome

Pressure injury prevention in COVID-19 patients with acute respiratory distress syndrome

Prevention of hospital-acquired pressure injury in COVID-19 patients in prone position, Intensive Crit Care Nurs. Advance online publication