- Email: [email protected]
The Surgical Optimal Mobility Score predicts mortality and length of stay in an Italian population of medical, surgical, and neurologic intensive care unit patients
The Surgical Optimal Mobility Score predicts mortality and length of stay in an Italian population of medical, surgical and neurologic ICU patients Simone Piva MD, Giancarlo Dora MD, Cosetta Minelli MD, PhD, Mariachiara Michelini MD, Fabio Turla, Stefania Mazza, Patrizia D’ Ottavi MD, D. Pharm., Ingrid Moreno-Duarte MD, Caterina Sottini MD, Matthias Eikermann MD, PhD, Nicola Latronico MD PII: DOI: Reference:
S0883-9441(15)00434-7 doi: 10.1016/j.jcrc.2015.08.002 YJCRC 51914
To appear in:
Journal of Critical Care
Please cite this article as: Piva Simone, Dora Giancarlo, Minelli Cosetta, Michelini Mariachiara, Turla Fabio, Mazza Stefania, D’ Ottavi Patrizia, Moreno-Duarte Ingrid, Sottini Caterina, Eikermann Matthias, Latronico Nicola, The Surgical Optimal Mobility Score predicts mortality and length of stay in an Italian population of medical, surgical and neurologic ICU patients, Journal of Critical Care (2015), doi: 10.1016/j.jcrc.2015.08.002
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
The Surgical Optimal Mobility Score predicts mortality and length of stay in an Italian
T
population of medical, surgical and neurologic ICU patients
RI P
Simone Piva, MD,1 Giancarlo Dora, MD,2 Cosetta Minelli, MD, PhD stat,3 Mariachiara Michelini, MD,2 Fabio Turla, Critical Care Nurse,1 Stefania Mazza Critical Care Nurse,1 Patrizia
SC
D' Ottavi, MD, D. Pharm.,4 Ingrid Moreno-Duarte, MD,5 Caterina Sottini, MD,2 Matthias
Department of Anesthesia, Critical Care Medicine and Emergency, Spedali Civili University
MA
1
NU
Eikermann, MD, PhD,5 and Nicola Latronico,MD1,4.
Hospital, Brescia, Italy 2
Department of Physical Medicine and Rehabilitation, Spedali Civili University Hospital,
University of Brescia, Italy
Department of Anesthesia, Critical Care, and Pain Medicine, Massachusetts General Hospital
AC
5
Department of Medical and Surgical Specialties, Radiological Sciences and Public Health,
CE
4
National Heart and Lung Institute, Imperial College London, London, England
PT
3
ED
Brescia, Italy
and Harvard Medical School, Boston, Massachusetts, USA
Corresponding author Prof. Nicola Latronico Department of Anesthesia, Critical Care Medicine and Emergency Spedali Civili University Hospital Piazzale Ospedali Civili, 1 – 25123 Brescia, Italy Tel. +39-030 3995 561 (secretary)/764 (ICU) Fax +39-030 3995 779 Email [email protected]
1
ACCEPTED MANUSCRIPT
ABSTRACT
RI P
T
Purpose
We validated the Italian version of Surgical Optimal Mobility Score (SOMS) and evaluated its
NU
mortality in a mixed population of ICU patients.
SC
ability to predict intensive care unit (ICU) and hospital length of stay (LOS), and hospital
MA
Materials and Methods
We applied the Italian version of SOMS in a consecutive series of prospectively enrolled, adult
ED
ICU patients. SOMS level was assessed by ICU nurses twice a day, and twice a week by an expert mobility team (EMT). Zero-truncated Poisson regression was used to identify predictors
PT
for ICU and hospital LOS, and logistic regression for hospital mortality. All models were
AC
Results
CE
adjusted for potential confounders.
Of 98 patients recruited, 19 (19.4%) died in hospital, of whom 17 without and 2 with improved mobility level achieved during the ICU. SOMS improvement was independently associated with lower hospital mortality (OR: 0.07; 95% CI: 0.01-0.42) but increased hospital LOS (OR: 1.21; 95% CI: 1.10-1.33). A higher first morning SOMS on ICU admission, indicating better mobility, was associated with lower ICU and hospital LOS (rate ratio: 0.89, 95% CI: 0.80-0.99, and 0.84, 95% CI: 0.79-0.89, respectively).
2
ACCEPTED MANUSCRIPT
Conclusions
T
The first morning SOMS on ICU admission predicted ICU and hospital LOS in a mixed
RI P
population of ICU patients. SOMS improvement was associated with reduced hospital mortality
SC
but increased hospital LOS suggesting the need of optimizing hospital trajectories after ICU
MA
NU
discharge.
ED
Word count. Abstract: 224 words. Body text: 3,628ß. Number of figures: 2. Number of
PT
supplemental figures: 1. Number of tables: 6. Number of references: 35.
AC
neurologic patients.
CE
Key words: early mobilization; rehabilitation; safety; mortality; validation; surgical patients;
3
ACCEPTED MANUSCRIPT
Introduction
RI P
T
Early mobilization during the intensive care unit (ICU) stay has been advocated to mitigate the effects of muscle weakness[1-5]. Early rehabilitation includes a spectrum of
SC
interventions ranging from passive motion exercises to ambulation and to the use of novel technologies such as cycle-ergometry and transcutaneous electrical muscle stimulation[6].
NU
Although some patients are unable to progress through all the different steps of mobilization,
MA
even minimal motor activities can reduce muscle weakness and wasting. The beneficial effects of early mobilization in the medical ICU have been reported in
ED
several studies[1, 2]. Although not free of risks[3, 5, 7], its safety profile is reported to be good
PT
with low rates of complications even in patients who are traditionally not mobilized, such as those with femoral vein or artery catheters [8-10]. Different strategies have been proposed to
CE
modify potential barriers to early mobilization, such as changing vascular catheter location,
AC
careful scheduling of procedures, and improved sedation management [11, 12]. The Surgical Optimal Mobilization Score (SOMS), a five-point numerical rating scale to guide goal-directed early mobilization therapy, has been demonstrated to be a predictor of mortality as well as ICU and hospital length of stay (LOS) in a surgical ICU population[13]. In their original study, Kasotakis and colleagues studied the reliability of SOMS score in 113 functionally independent surgical ICU patients by comparing SOMS assigned by with those assigned by an expert mobility team, and they found an excellent agreement between the two.
4
ACCEPTED MANUSCRIPT
The authors also found that higher SOMS scores, indicating better mobility, were associated with
T
lower mortality and hospital and ICU LOS[13].
RI P
In this prospective cohort study, we assessed if SOMS can predict ICU and hospital LOS
SC
and hospital mortality in an Italian population of adult medical, surgical and neurologic ICU patients. We also evaluated the inter-rater reliability of the Italian version of SOMS and the
NU
safety of SOMS-guided mobilization in this mixed ICU population.
MA
Materials and Methods
The SOMS is an algorithm for goal-directed early mobilization in the ICU. It contains a
ED
numerical rating scale from 0 to 4 to quantify the patient's mobilization capacity. SOMS 0
PT
indicates that no mobilization should be considered since deemed to be futile, as for patients in terminal unstable clinical condition such as those with intracranial hypertension or severe
CE
systemic hemodynamic and respiratory insufficiency. SOMS 1 indicates that the patient can
AC
receive passive range of motion exercise while in bed and SOMS 2 that the patient can be sitting up in bed. SOMS 3 indicates that the patient is able to stand with or without assistance, and SOMS 4 is assigned to patients able to ambulate[13]. Development of the Italian version of the SOMS As a first step to use SOMS in our ICU, we provided an Italian translation of the original English version following the recommendations for a comprehensive multistep process for translating, adapting and cross-validating instruments (supplemental figure 1)[14]. Two qualified medical
5
ACCEPTED MANUSCRIPT
doctors whose native language was Italian, fluent in English and with knowledge in early
T
rehabilitation in the ICU, independently translated into Italian the original version of the SOMS
RI P
score, the associated instructions, and the drawings’ text. Thereafter, a consensus meeting was held to agree on a fully comprehensible and accurate Italian translation consistent with the
SC
original English text. The draft was back translated into English and compared with the original
NU
to develop the final Italian translation.
MA
Study design and setting
The study was a prospective observational study conducted at the general and neurologic ICU of
ED
the Department of Anesthesia, Critical Care and Emergency of the Spedali Civili of Brescia, a large regional university-affiliated hospital. The ICU has 10 beds, 6 general and 4 neurologic.
PT
The daytime staffing of the unit consists of one medical coordinator, one attending physician,
CE
two residents (4th and 5th-year residents of the School of Specialty in Anesthesia and Critical Care Medicine), and six critical care nurses. The night shift team consists of one attending
AC
physician, one resident, and four nurses. Physical therapists are not dedicated exclusively to the ICU, and provide general and respiratory physical therapy for 5 days a week based on a physiatrist-activated written protocol. The study was approved by the local Ethics Committee (Comitato Etico Provinciale di Brescia, June 17, 2013, Ref. N. 1383). Detailed written information was provided to the patients and family members about the study, and written informed consent to participate to the study was obtained from the patient whenever possible. In case of altered consciousness, the Ethics
6
ACCEPTED MANUSCRIPT
Committee waived the requirement for consent, as in Italy relatives are not regarded as legal
T
representatives of the patient in the absence of a formal designation. Written informed consent
RI P
was subsequently requested from all surviving patients as soon as they regained their mental competency. The study was conducted according to the principles expressed in the Declaration
SC
of Helsinki.
NU
Patients were eligible if they were 18 years or older and were expected to stay in the ICU for at least 72 hours. Patients were defined as medical, surgical or neurological (patient's class)
MA
according to the reason for ICU admission. Patients with predicted ICU stay of less than 72 hours, admitted for post-surgery monitoring or with unstable spine or in terminal condition were
PT
Patient management
ED
excluded.
CE
All patients were managed following the early-goal directed therapy guidelines [15] and a goal-
AC
directed sedation protocol aimed at minimizing the use of sedatives through daily interruption[16]. In non-neurologic patients, daily interruption of sedation was part of the “Awakening and Breathing Coordination, Delirium monitoring/management, and Early exercise/mobility” (ABCDE) bundle. SOMS assessment of the level of mobility anticipated to be accomplished during the morning shift was performed at fixed timed (see Study procedure) and not necessarily after awakening. However, the morning shift nurse integrated the entire information obtained from the night shift nurse with the actual patient’s condition in order to assign SOMS scores on a solid clinical base [17].
7
ACCEPTED MANUSCRIPT
Critically ill neurologic patients had continuous monitoring of cerebral hemodynamics, including
T
intracranial pressure, cerebral perfusion pressure and cerebro-vascular autoregulation
RI P
monitoring, according to pre-defined protocol [18]. For multimodal data acquisition, we used the Intensive Care Monitoring software system (ICM+, University of Cambridge, UK) running on
SC
bedside laptop computers[18]. Neurological severity was graded according to the Glasgow Coma
NU
Scale [19] and the Full Outline of UnResponsiveness scale [20]. Continuous sedation and analgesia with propofol and fentanyl was gradually reduced and then interrupted if the
MA
neurological condition, the systemic and cerebral hemodynamics and brain CT findings stabilized according to the neurointensivist in charge. Initial repeated interruptions of sedation up
ED
to 3 times a day were followed by definitive interruption if the patient could tolerate it with no
PT
excessive stress reactions. This practice of early interruption of sedation has been in place for
Staff training
CE
several years in our unit, and has been demonstrated to be safe also by other centers[21].
AC
From April to June 2013 all ICU nurses were trained by study members to apply the SOMS in simulated cases and during routine care. The ICU staff was also provided with written instructions for proper SOMS scoring, and they were involved in 3 educational meetings devoted to present the potential benefits of early mobilization in the ICU. From July 2013 to October 2013 we prospectively recruited a consecutive series of critically ill patients admitted to our ICU.
8
ACCEPTED MANUSCRIPT
Study procedures
T
Study investigators screened all ICU admissions daily in the morning to identify those patients
RI P
fulfilling the inclusion criteria. The early mobilization program was started on the day after
SC
enrollment.
The SOMS level was assessed twice a day for each patient by two nurses during the morning
NU
shift at least 30 minutes after handover from the night shift nurse (morning SOMS) and then
MA
after the lunch break (afternoon SOMS). The two nurses' evaluations were separated from one another by a maximum of 30 minutes to minimize the effect of clinical fluctuation. The morning
ED
SOMS level was defined as the level of mobility anticipated to be accomplished during the morning shift. The level of mobilization effectively reached was defined as achieved SOMS. All
PT
nurses performed the assessments independently, and were blinded to the other’s assessment
CE
both in the morning and afternoon sessions. In addition, the nurses also recorded any barrier or complication (see next section) related to mobilization[22]. An expert mobility team (EMT)
AC
including an intensivist and a rehabilitation physician assessed the patients twice a week on Tuesday and Thursday morning to predict mobilization capacity using their expertise and the SOMS algorithm. The team, which was blinded to the morning SOMS predicted by the nurses, assigned a new independent SOMS score (EMT-SOMS). Progress in mobility goal (delta SOMS) for every day followed the SOMS algorithm (figure 1). A delta SOMS of 1 or 0 indicated progress or, respectively, lack thereof in mobility goals. Dedicated study members
9
ACCEPTED MANUSCRIPT
collected all SOMS scores and recorded the achieved SOMS, as well as other relevant
T
information.
RI P
Safety monitoring of SOMS-guided mobilization
SC
To ensure the safety of the subjects, each day the nursing staff monitored any mobilization-
NU
associated adverse event (AE) that occurred during or within 30 minutes after mobilization. We classified as serious adverse events (SAE),any unintended injury that was due at least in part
MA
to incorrect mobilization and that exposed the patient to an increased risk of death, measurable disability or intervention to prevent permanent impairment or damage[24]. Since most SAE are
ED
preceded by clinically observable warning signs[24], we defined the following medical events to
PT
withheld mobility interventions: a decline in hemodynamic (mean arterial pressure < 65 mmHg or systolic blood pressure < 90 mmHg) or respiratory status (SpO2 <88%); hypertensive crisis
CE
(systolic blood pressure >180 mmHg); need for administration of a new vasopressor agent; need
AC
to increase the positive end-expiratory pressure on the ventilator or a change from spontaneous breathing or continuous positive airway pressure to assist control mode once in a weaning mode; symptoms of acute myocardial ischemia, or cardiac arrhythmias requiring anti-arrhythmic agent; accidental removal of a device (airways and cardiac devices, chest tubes, vascular access, wound or dressing); severe pain (numeric rating scale, NRS, of 4 or more in collaborative patients and need for analgesics in non-collaborative patients); fall, defined as an unplanned descent of the patient to the floor, and classified as with or without injury according to American Nurses
10
ACCEPTED MANUSCRIPT
Association, National Database of Nursing Quality Indicators [25]. Reported AEs and associated
T
complications were prospectively recorded by dedicated study members.
RI P
Data presentation and statistical analysis
SC
We expressed continuous variables as means (SD) or as medians (interquartile range, IQR) as appropriate. Discrete variables, including AE and other safety variables, were expressed as
NU
counts (percentage).
MA
We evaluated if SOMS can predict ICU and hospital LOS hospital and hospital mortality in order to replicate the results of the original SOMS study [13] in our mixed ICU population.
ED
Based on the original study [13], we anticipated an inverse correlation between the first morning
PT
SOMS and the ICU LOS of r ≥ 0.35. We calculated that a sample size of 98 patients would provide 80% power to detect such an association between the first morning SOMS and the ICU
CE
LOS with a two-sided α of 0.05.
AC
For ICU and hospital LOS, we used a zero-truncated Poisson regression model with rate ratios to identify predictors predictors. The following predicting variables were considered: age, first morning SOMS after ICU admission, delta SOMS, SAPS II score, comorbidity index, patient’s class (neurological versus others)[26], and use of vasopressors and renal replacement therapy. For the adjusted analysis, variables were included in the model if they were statistically significant in unadjusted analysis or if they were clinically relevant.
11
ACCEPTED MANUSCRIPT
For hospital mortality, potential predictors were identified using unadjusted logistic regression
T
with odds ratios (OR). As for LOS, for the adjusted analysis variables were included in the
RI P
model if they were statistically significant in unadjusted analysis or were clinically relevant. Overall model evaluation against the null model was obtained using the likelihood ratio test.
SC
Model calibration was assessed using the Hosmer-Lemeshow test, while the Area Under
NU
Receiver Operating Characteristic curve (AUROC) was used to assess discrimination. AUROC value >0.70, >0.80 and >0.90 were considered as acceptable, excellent and outstanding,
MA
respectively [27]. For each outcome we also estimated the variance explained by the model, R2. To assess SOMS inter-rater reliability in pairwise comparisons, we compared: a) SOMS
ED
assigned by the two nurses in the morning (morning SOMS), and b) in the afternoon (afternoon
PT
SOMS), and c) morning SOMS by the two nurses (the median value) versus EMT-SOMS. Interrater reliability was investigated using Spearman rho correlation coefficients with corresponding
CE
p-values among the three raters, and the percentage agreement. Correlation coefficients below
AC
0.40 were defined poor, between 0.41to 0.60 moderate, between 0.61 to 0.80 good, and above 0.80 very good [28].
For all analyses, a p-value of less than 0.05 was considered to indicate statistical significance. Statistical analyses were performed using STATA 13 (Stata Corp LP, College Station, Texas, USA).
12
ACCEPTED MANUSCRIPT
Results
RI P
T
During the study period, 187 patients were admitted to the ICU, 89 were excluded (41 not critically ill and admitted for postoperative monitoring only; 32 with an ICU stay of less than 72
SC
hours; 9 younger than 18 years; 7 in terminal condition), and 98 were recruited in the study. Demographic characteristics, admission diagnoses, patient severity and outcome are presented in
NU
table 1.
MA
In total, we obtained 854 morning SOMS evaluations: SOMS 0 was assigned in 41 cases (4.8%), SOMS 1 in 500 (58.5%), SOMS 2 in 230 (26.9%), SOMS 3 in 65 (7.6%) and SOMS 4 in 18
ED
(2.1%) cases.
PT
In 61 occasions (7.1%), the achieved SOMS was lower than morning SOMS, due to: need for sedation (n=24 evaluations out of 854), agitation (n=13), presence of femoral catheter (n=11),
CE
surgery (n=5), arterial hypotension (n=3), absence of resources for mobilization (n=3), patient
AC
refusal (n=3), delirium (n=3) and mild oxygen desaturation (n=1). There were 10 AEs (1.2% of all mobilization sessions), of which 1 (0.1%) was a SAE of hypotension requiring a vasopressor agent in a 65-year old patient with abdominal septic shock and a history of cardiac arrhythmias. This patient improved progressively and was safely discharged to a rehabilitation center, but died upon initiation of rehabilitation due to fatal ventricular dysrhythmia. Of the other 9 cases (4 of mild hypotension, 3 of mild pain and 2 falls from chair without injury), only 2 (0.2%) occurred in neurologic patients (2 falls).
13
ACCEPTED MANUSCRIPT
The median ICU LOS was 8 days (IQR= 4.0-13.5 days). In the unadjusted analyses, ICU LOS
T
was inversely associated with the first morning SOMS and delta SOMS (the higher the first
RI P
morning SOMS and delta SOMS the lower the ICU LOS), and directly associated with SAPS II scores and comorbidity index (table 2). In the adjusted analysis, ICU LOS was inversely
SC
associated with first morning SOMS (an improvement of 1 point of the SOMS score decreased
NU
the ICU LOS by 11.1%) and directly associated with SAPS II and comorbidity index (table 3). Hospital LOS had a median of 20 days (IQR= 12.0-33.5 days). In the unadjusted analyses it was
MA
inversely associated with first morning SOMS and patient's class (hospital LOS was lower in non-neurologic patients), and directly associated with delta SOMS (patients with improved
ED
mobility level during the ICU stay had longer hospital LOS), SAPS II scores and comorbidity
PT
index (table 2). In the adjusted analysis, hospital LOS was inversely associated with first morning SOMS (an improvement of 1 point of the SOMS score decreased hospital LOS by 16%)
CE
and patient's class, and directly associated with delta SOMS (an improved mobility level
AC
increased the hospital LOS by 20.7%) (table 3). Of 98 patients, 19 (19.4%) died in hospital, of whom 17 without and 2 with a progression in mobility goals. In the unadjusted analyses the hospital mortality was inversely associated with delta SOMS and directly associated with SAPS II (table 4). Both remained significant in the adjusted analysis (table 5). The fit of the multivariable model was good (Hosmer-Lemeshow test p-value: 0.97), and its overall accuracy in predicting mortality was excellent, with an AUROC of 0.89 (figure 2).
14
ACCEPTED MANUSCRIPT
Spearman rho correlation coefficients were very good for all comparisons, and the percentage
RI P
T
agreement between SOMS raters was always higher than 80% (table 6).
SC
Discussion
NU
We translated the original English version of SOMS in Italian using a rigorous methodology and
MA
we applied it in a mixed population of unselected medical, surgical and neurologic ICU patients. SOMS showed reliably when used in routine clinical practice to guide early mobilization. SOMS
ED
scores were applied by ICU staff with excellent inter-rater reliability, comparable to that reported by the developers of the scale. Reliability was excellent both in nurse-to-nurse and nurse-to-
PT
expert comparisons. The SOMS level of mobility predicted by nurses on the first morning after
CE
ICU admission was independently associated with ICU and hospital LOS. Improvement in the level of mobility achieved during the ICU stay was independently associated with reduced
AC
hospital mortality, but also with prolonged hospital LOS. These data support the hypothesis that SOMS has construct validity and that is a valid algorithm to select the optimal patient's mobility level. Moreover, they confirm the results of the original study showing that SOMS at ICU admission predicted ICU and hospital LOS[13]. SOMS-standardized mobilization was safely implemented by ICU nurses after an adequate training with application of SOMS in simulated and real cases. Absence of sufficient resources was the cause of failed mobilization in only 3 of 854 evaluations, indicating that SOMS can be
15
ACCEPTED MANUSCRIPT
implemented in routine ICU care with no supplemental cost or need for specialized equipment.
T
AE occurred in 1.2 percent of all mobilization sessions, which compares well with other studies
RI P
where figures varied from 0% to 22% [9]. The AE rate was excellent also in acutely ill neurologic patients, of whom one third could sit up in bed, stand or ambulate during the ICU
SC
stay, indicating that early mobilization can be safely implemented in this category of patients,
NU
provided that cerebral and systemic physiology are stabilized. Indeed, the neurocritical care can be an ideal setting for this intervention because immobility is common consequence of
MA
neurological dysfunction. Titsworth and colleagues reported similar findings in a population of 93 neurocritical care patients mobilized according to an 11-step Progressive Upright Mobility
ED
Protocol (PUMP) Plus algorithm [32]; however, PUMP Plus has not been validated and requires
PT
the acquisition of specialized equipment and dedicated personnel. Patients with pre-existing physical dysfunction may benefit less from early mobilization, but unlike the original study, we
CE
did not exclude patients with reduced functional independence (Barthel index score 70) before
AC
ICU admission. This may increase the generalizability of our results, further reinforcing that SOMS can predict LOS and mortality in unselected ICU patient populations. ICU LOS was reduced in patients with higher SOMS scores on the first morning after ICU admission and in those with less severe disease and burden of comorbid disease. Morning SOMS as a mobility level predicted to be accomplished during the morning shift can be considered itself a descriptor of patient’s severity. Clinician estimates of patient’s outcome in the ICU are often accurate, reflecting a great deal of experience [33]. This finding is in line with the result of the original study showing that SOMS predicts ICU LOS as well as the APACHE score [13]. More
16
ACCEPTED MANUSCRIPT
than 60 percent of our patients were acutely ill neurologic patients, and hence, our results also
T
confirm those of Titsworth showing a reduced duration of the ICU stay upon implementation of
RI P
an early mobility program among neurocritical care unit patients [32].
SC
As for ICU LOS, hospital LOS was reduced in patients with higher first morning SOMS, in those with less severe disease and burden of comorbid disease, and in those with a surgical or medical
NU
cause of ICU admission compared to neurologic patients. Unexpectedly, hospital LOS was increased in patients with improvement of the mobility level during the ICU stay. Reasons for
MA
this result remain speculative because we did not plan specific follow-up investigations after ICU discharge. In the original study enrolling a homogeneous cohort of surgical ICU patients, a
ED
higher SOMS score was associated with reduced hospital stay, but delta SOMS was not
PT
evaluated[13]. Delta SOMS has a completely different meaning compared to the first morning SOMS, representing the improvement in mobility goals achieved during the ICU stay and not a
CE
proxy of disease's severity. We speculate that transfer of ICU patients to several different wards,
AC
each with specific discharge policy, may have influenced the results of our study. Future studies should consider the optimization of post-ICU trajectories within the acute-care hospital in order to properly evaluate the multidimensional aspects influencing the hospital LOS. This study is the first to demonstrate an independent association between improvement of mobility during the ICU stay and reduced mortality. Other studies have demonstrated improved functional independence at hospital discharge with a strategy of whole-body rehabilitation compared to standard care, but mortality was unchanged [1]. Although promising, this result
17
ACCEPTED MANUSCRIPT
must be interpreted with caution. Due to the observational nature and the small sample size of
T
the study, we could only consider a limited number of covariates such as the disease severity and
RI P
comorbidity in our multivariable model, while mortality in critically ill patients is influenced by multiple complex factors. Predictive modeling is difficult when considering the numerous
SC
clinical elements that occur after critical illness and their interplay[34, 35].
NU
Future efficacy trials with sufficient power and longer-term outcomes are needed to establish if SOMS-guided early mobilization may reduce mortality and functional disability in survivors of
ED
MA
critical illness.
PT
Conclusion
CE
Use of the SOMS algorithm to guide optimal patient's mobility level was safely implemented by ICU nurses in the routine care of a mixed population of medical, surgical and neurologic ICU
AC
patients after adequate training.
Inter-rater reliability of SOMS scores was excellent both in nurse-to-nurse and nurse-to-expert mobility team comparisons. Absence of sufficient resources was a rare cause of failed mobilization, indicating that SOMS can be implemented in routine ICU care with no supplemental cost or need for specialized equipment.
18
ACCEPTED MANUSCRIPT
Improvement in the level of mobility achieved during the ICU stay was independently associated
RI P
T
with reduced ICU length of stay and hospital mortality.
SC
Abbreviations
NU
AEs, adverse events during the ICU stay.
MA
EMT, expert mobility team.
LOS, length of stay.
PT
SAEs, serious adverse events.
ED
ICU, intensive care unit.
CE
SAPS II, Simplified Acute Physiology Score II.
AC
SOMS, Surgical Optimal Mobility Score.
Competing interests The authors have no competing interests to declare.
19
ACCEPTED MANUSCRIPT
Authors’ contributions
S. Piva, C. Minelli and N. Latronico drafted the manuscript.
RI P
T
S. Piva, M. Eikermann and N. Latronico were responsible for designing the study.
SC
S. Piva, N. Latronico, I. Moreno-Duarte, M. Eikermann and C. Minelli were responsible for data
NU
analysis and contributed to the revision of the manuscript.
simulated cases and during routine care.
MA
S. Piva, G. Dora and C. Sottini were responsible for training the ICU staff in using SOMS in
ED
S. Piva and G. Dora were responsible for organizing the expert mobility team. F. Turla and S. Mazza were responsible for providing continuous support to the critical care
PT
nurses in assigning SOMS score.
CE
S. Piva, F. Turla, S. Mazza provided written instructions for proper SOMS scoring, and
AC
organized the educational meetings on early ICU mobilization. S. Piva and N. Latronico translated the English version of SOMS in Italian and drafted the initial version of the paper.
M. Michelini and P. D'Ottavi were responsible for collecting and recording the SOMS scores, the achieved level of mobility, the clinical and demographic data and follow-up, and contributed to data analysis and interpretation. All authors read and approved the final version of the manuscript.
20
ACCEPTED MANUSCRIPT
1.
RI P
T
References
Schweickert WD, Pohlman MC, Pohlman AS, Nigos C, Pawlik AJ, Esbrook CL, et al.
SC
Early physical and occupational therapy in mechanically ventilated, critically ill patients: a randomised controlled trial. Lancet 2009; 373:1874-82
Burtin C, Clerckx B, Robbeets C, Ferdinande P, Langer D, Troosters T, et al. Early
NU
2.
MA
exercise in critically ill patients enhances short-term functional recovery. Critical care medicine 2009; 37:2499-505
Kayambu G, Boots R, Paratz J. Physical therapy for the critically ill in the ICU: a
ED
3.
systematic review and meta-analysis. Crit Care Med 2013; 41:1543-54 Eikermann M, Latronico N. What is new in prevention of muscle weakness in critically
PT
4.
5.
CE
ill patients? Intensive Care Med 2013; 39:2200-3 Parker A, Sricharoenchai T, Needham DM. Early Rehabilitation in the Intensive Care
AC
Unit: Preventing Physical and Mental Health Impairments. Current physical medicine and rehabilitation reports 2013; 1:307-14 6.
Berney S, Haines K, Denehy L. Physiotherapy in critical care in australia.
Cardiopulmonary physical therapy journal 2012; 23:19-25 7.
Needham DM, Davidson J, Cohen H, Hopkins RO, Weinert C, Wunsch H, et al.
Improving long-term outcomes after discharge from intensive care unit: report from a stakeholders' conference. Crit Care Med 2012; 40:502-9
21
ACCEPTED MANUSCRIPT
8.
Perme C, Nalty T, Winkelman C, Kenji Nawa R, Masud F. Safety and Efficacy of
9.
RI P
Observational Study. Cardiopulm Phys Ther J 2013; 24:12-7
T
Mobility Interventions in Patients with Femoral Catheters in the ICU: A Prospective
Sricharoenchai T, Parker AM, Zanni JM, Nelliot A, Dinglas VD, Needham DM. Safety
SC
of physical therapy interventions in critically ill patients: a single-center prospective evaluation
10.
NU
of 1110 intensive care unit admissions. J Crit Care 2014; 29:395-400 Fan E, Cheek F, Chlan L, Gosselink R, Hart N, Herridge MS, et al. An official American
MA
Thoracic Society Clinical Practice guideline: the diagnosis of intensive care unit-acquired weakness in adults. Am J Respir Crit Care Med 2014; 190:1437-46 Needham DM, Korupolu R, Zanni JM, Pradhan P, Colantuoni E, Palmer JB, et al. Early
ED
11.
PT
physical medicine and rehabilitation for patients with acute respiratory failure: a quality improvement project. Archives of physical medicine and rehabilitation 2010; 91:536-42 Engel HJ, Tatebe S, Alonzo PB, Mustille RL, Rivera MJ. Physical therapist-established
CE
12.
AC
intensive care unit early mobilization program: quality improvement project for critical care at the University of California San Francisco Medical Center. Physical therapy 2013; 93:975-85 13.
Kasotakis G, Schmidt U, Perry D, Grosse-Sundrup M, Benjamin J, Ryan C, et al. The
surgical intensive care unit optimal mobility score predicts mortality and length of stay. Critical care medicine 2012; 40:1122-8 14.
Wild D, Grove A, Martin M, Eremenco S, McElroy S, Verjee-Lorenz A, et al. Principles
of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes (PRO) Measures: report of the ISPOR Task Force for Translation and Cultural
22
ACCEPTED MANUSCRIPT
Adaptation. Value in health : the journal of the International Society for Pharmacoeconomics and
Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving
RI P
15.
T
Outcomes Research 2005; 8:94-104
Sepsis Campaign: international guidelines for management of severe sepsis and septic shock,
Girard TD, Kress JP, Fuchs BD, Thomason JW, Schweickert WD, Pun BT, et al.
NU
16.
SC
2012. Intensive Care Med 2013; 39:165-228
Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically
MA
ventilated patients in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial. Lancet 2008; 371:126-34
Vasilevskis EE, Ely EW, Speroff T, Pun BT, Boehm L, Dittus RS. Reducing iatrogenic
ED
17.
PT
risks: ICU-acquired delirium and weakness--crossing the quality chasm. Chest 2010; 138:122433
Rasulo FA, Girardini A, Lavinio A, De Peri E, Stefini R, Cenzato M, et al. Are optimal
CE
18.
AC
cerebral perfusion pressure and cerebrovascular autoregulation related to long-term outcome in patients with aneurysmal subarachnoid hemorrhage? J Neurosurg Anesthesiol 2012; 24:3-8 19.
Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical
scale. Lancet 1974; 2:81-4 20.
Wijdicks EF, Bamlet WR, Maramattom BV, Manno EM, McClelland RL. Validation of a
new coma scale: The FOUR score. Ann Neurol 2005; 58:585-93
23
ACCEPTED MANUSCRIPT
21.
Skoglund K, Hillered L, Purins K, Tsitsopoulos PP, Flygt J, Engquist H, et al. The
22.
RI P
with severe traumatic brain injury. Neurocrit Care 2014; 20:413-26
T
neurological wake-up test does not alter cerebral energy metabolism and oxygenation in patients
Leditschke IA, Green M, Irvine J, Bissett B, Mitchell IA. What are the barriers to
Brennan TA, Localio AR, Leape LL, Laird NM, Peterson L, Hiatt HH, et al.
NU
23.
SC
mobilizing intensive care patients? Cardiopulmonary physical therapy journal 2012; 23:26-9
Identification of adverse events occurring during hospitalization. A cross-sectional study of
MA
litigation, quality assurance, and medical records at two teaching hospitals. Ann Intern Med 1990; 112:221-6
Jones DA, DeVita MA, Bellomo R. Rapid-response teams. N Engl J Med 2011; 365:139-
ED
24.
PT
46
Currie LM. Fall and injury prevention. Annu Rev Nurs Res 2006; 24:39-74
26.
Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying
CE
25.
40:373-83
AC
prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;
27.
David W. Hosmer, Jr. Applied Logistic Regression, 3rd Edition.
28.
David W. Hosmer J, Stanley Lemeshow, Rodney X. Sturdivant: Applied Logistic
Regression, Third Edition, 2013 29.
Morris PE, Goad A, Thompson C, Taylor K, Harry B, Passmore L, et al. Early intensive
care unit mobility therapy in the treatment of acute respiratory failure. Crit Care Med 2008; 36:2238-43
24
ACCEPTED MANUSCRIPT
30.
Elliott D, McKinley S, Alison J, Aitken LM, King M, Leslie GD, et al. Health-related
T
quality of life and physical recovery after a critical illness: a multi-centre randomised controlled
31.
RI P
trial of a home-based physical rehabilitation program. Crit Care 2011; 15:R142 Skinner EH, Berney S, Warrillow S, Denehy L. Development of a physical function
SC
outcome measure (PFIT) and a pilot exercise training protocol for use in intensive care. Critical
NU
care and resuscitation : journal of the Australasian Academy of Critical Care Medicine 2009; 11:110-5
Titsworth WL, Hester J, Correia T, Reed R, Guin P, Archibald L, et al. The effect of
MA
32.
increased mobility on morbidity in the neurointensive care unit. Journal of neurosurgery 2012;
Rocker G, Cook D, Sjokvist P, Weaver B, Finfer S, McDonald E, et al. Clinician
PT
33.
ED
116:1379-88
predictions of intensive care unit mortality. Crit Care Med 2004; 32:1149-54 Latronico N. Prediction is very difficult, especially about the future*. Crit Care Med
35.
AC
2015; 43:505-6
CE
34.
Latronico N, Herridge MS. Unraveling the myriad contributors to persistent diminished
exercise capacity after critical illness. Intensive care medicine 2015;
25
ACCEPTED MANUSCRIPT
Figure 1. Surgical Optimal Mobility Score.
SOMS 0: No activity
SOMS 1: PROM, upright in bed
SOMS 2: Sitting up
SOMS 3: Standing
SOMS 4: Ambulating
MA
NU
SC
1. Stable spine, no SCI. 2. ICP < 20 mmHg 3. Not a moribund patient 1. Follow simple commands. 2. No open spinal drains, EVD, peritoneum, chest. 3. No femoral CVVH lines.
RI P
Reproduced with permission of the Editor Lippincott Williams & Wilkins.
T
CVVH, continuous veno-venous hemofiltration; EVD, extracranial ventricular drains; ICP, intracranial pressure; PROM, passive range of motion; SCI, spinal cord injury.
1. Stand twice with minimal assist. 2. Steps in place with minimal assist.
AC
CE
PT
ED
1. Bilateral quadriceps strength 3/5 or more 2. Sits without support 3. No weight -bearing restrictions.
26
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
Figure 2. Receiver-operating characteristic curve for hospital mortality, multivariable adjusted regression.
27
ACCEPTED MANUSCRIPT
Table 1. Demographic characteristics of participants, admission diagnoses and outcome. ICU, intensive care unit. SAPS II, Simplified Acute Physiology Score.
RI P
T
Measure 61.5 (16.4) 67 (68.3%)
PT
CE
Mechanical ventilation [number of patients (percentage)] SAPS II score [mean (SD)] Vasopressor therapy [number of patients (percentage)] Renal replacement therapy [number of patients (percentage)] ICU length of stay, days [mean (SD)] Neurologic patients Surgical patients Medical patients Hospital length of stay, days [mean (SD)] Neurologic patients Surgical patients Medical patients Hospital mortality [number (percentage)]
AC
60 (61.2) 18 15 7 3 2 2 13 15 (15.3) 4 2 4 5 23 (23.5) 11 8 4
SC NU
ED
MA
Variables Age, years [mean (SD)] Male [number (percentage)] Diagnosis at ICU admission [number (percentage)] Neurological diseases head trauma subarachnoid hemorrage intracerebral hematoma encephalitis seizure/epilepsy post-cardiac arrest encephalopathy other diagnoses Surgical diseases major vascular surgery abdominal surgery polytrauma other diagnoses Medical diseases sepsis/septic shock acute respiratory distress syndrome other diagnoses
76 (77.5%) 41.1 (15.7) 29 (29.5) 6 (6.1) 9.5 (6.4) 9.8 (5.7) 11.3 (8.6) 7.6 (6.5) 25.5 (23.7) 24.0 (26.4) 35.5 (17.7) 23 (17.4) 19 (19.3)
28
ACCEPTED MANUSCRIPT
Table 2. Predictors of the intensive care unit (ICU) and hospital length of stay, unadjusted analysis.
Variables
IRR
95% C.I.
p
SC
ICU length of stay
RI P
T
SOMS, surgical optimal mobility score. IRR, incidence risk ratio, is a hazard ratio and is interpreted in the same way. SAPS II, Simplified Acute Physiology Score. Patient's class indicates the cause of ICU admission (medical, surgical, neurological), with neurological patients as the reference class.
0.904
0.825-0.991
0.032
Delta SOMS
0.820
0.716-0.939
<0.004
Patient's class
1.072
0.9304-1.236
0.334
SAPS II score
1.015
1.010-1.019
<0.001
Comorbidity Index
1.212
MA
NU
First morning SOMS
1.129-1.301
<0.001
Hospital length of stay 0.869
Delta SOMS
1.086
Patient's class
0.855
SAPS II score Comorbidity Index
0.822-0.920
ED
First morning SOMS
<0.001 0.051
0.787-0.931
<0.001
1.009
1.007-1.012
<0.001
1.274
1.221-1.329
<0.001
AC
CE
PT
0.999-1.181
29
ACCEPTED MANUSCRIPT
Table 3. Predictors of the intensive care unit (ICU) and hospital length of stay (LOS), multivariable adjusted analysis.
RI P
T
SOMS, surgical optimal mobility score. IRR, incidence risk ratio, is a hazard ratio and is interpreted in the same way. SAPS II, Simplified Acute Physiology Score. Patient's class indicates the cause of ICU admission (medical, surgical, neurological), with neurological patients as the reference class.
Variable
SC
R2= 0.1048 (ICU LOS) and 0.0837 (hospital LOS).
IRR
95% C.I.
p
SAPS II
1.011
Comorbidity Index
1.112
Constant
5.983
Variable
IRR
0.802-0.986
0.025
1.006-1.016
<0.001
1.022-1.209
0.013
4.402-8.131
<0.001
95% C.I.
p
MA
0.889
ED
First morning SOMS
NU
ICU length of stay
Hospital length of stay
1.207
First morning SOMS Patient's class
Constant
0.840
0.789-0.893
<0.001
0.908
0.834-0.989
0.027
1.005
1.002-1.009
<0.001
1.250
1.190-1.313
<0.001
18.640
15.522-22.384
<0.001
AC
Comorbidity Index
<0.001
CE
SAPS II
1.100-1.325
PT
Delta SOMS
30
ACCEPTED MANUSCRIPT
Table 4. Predictors of hospital mortality, unadjusted regression analysis.
T
SOMS, surgical optimal mobility score. OR, odds ratio. SAPS II, Simplified Acute Physiology Score. CVVH, Continuous Venous-Venous Hemofiltration. Patient's class indicates the cause of ICU admission (medical, surgical, neurological), with neurological patients as the reference class.
95% C.I.
First morning SOMS
0.995
0.494-2.007
0.991
Delta SOMS
0.066
0.014-0.308
0.001
Patient's class
0.363
0.130-1.018
SAPS II score
1.077
1.032-1.124
0.001
Comorbidity Index
1.538
Vasopressors
1.306
CVVH
1.906
0.887-2.667
0.125
0.426-4.003
0.641
0.320-11.353
0.479
NU
0.054
AC
CE
PT
ED
MA
p
SC
RI P
OR
31
ACCEPTED MANUSCRIPT
Table 5. Predictors of hospital mortality, multivariable adjusted analysis. SOMS, surgical optimal mobility score. SAPS II, Simplified Acute Physiology Score. Patient's class indicates the cause of ICU admission, either medical, surgical or neurological. OR= Odds ratio. C.I.= Confidence Interval.
RI P
T
R2= 0.2861.
OR
95% C.I.
Delta SOMS
0.066
0.010-0.417
SAPS II
1.071
1.007-1.140
0.030
Constant
0.222
0.010-4.689
0.033
P 0.004
AC
CE
PT
ED
MA
NU
SC
Variable
32
ACCEPTED MANUSCRIPT
Table 6. Inter-rater reliability of Surgical Optimal Mobility Score (SOMS).
Spearman rho
P-value
0.982
<0.001
0.984
<0.001
80.32%
0.813
<0.001
PT
RI P
Percentage of agreement
T
EMT, expert mobility team. *We used the median morning SOMS assigned by two critical care nurses
Raters
94.39%
0.981
<0.001
Agreement between the afternoon SOMS assigned by two critical care nurses
98.55%
0.984
<0.001
84.40%
0.840
<0.001
SC
All patients 92.76%
NU
Agreement between the morning SOMS assigned by two critical care nurses
ED
Agreement between morning SOMS* and EMT-SOMS
98.30%
MA
Agreement between the afternoon SOMS assigned by two critical care nurses
Neurologic patients
AC
CE
Agreement between the morning SOMS assigned by two critical care nurses
Agreement between morning SOMS* and EMT-SOMS
33
S0883-9441(15)00434-7 doi: 10.1016/j.jcrc.2015.08.002 YJCRC 51914
To appear in:
Journal of Critical Care
Please cite this article as: Piva Simone, Dora Giancarlo, Minelli Cosetta, Michelini Mariachiara, Turla Fabio, Mazza Stefania, D’ Ottavi Patrizia, Moreno-Duarte Ingrid, Sottini Caterina, Eikermann Matthias, Latronico Nicola, The Surgical Optimal Mobility Score predicts mortality and length of stay in an Italian population of medical, surgical and neurologic ICU patients, Journal of Critical Care (2015), doi: 10.1016/j.jcrc.2015.08.002
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
The Surgical Optimal Mobility Score predicts mortality and length of stay in an Italian
T
population of medical, surgical and neurologic ICU patients
RI P
Simone Piva, MD,1 Giancarlo Dora, MD,2 Cosetta Minelli, MD, PhD stat,3 Mariachiara Michelini, MD,2 Fabio Turla, Critical Care Nurse,1 Stefania Mazza Critical Care Nurse,1 Patrizia
SC
D' Ottavi, MD, D. Pharm.,4 Ingrid Moreno-Duarte, MD,5 Caterina Sottini, MD,2 Matthias
Department of Anesthesia, Critical Care Medicine and Emergency, Spedali Civili University
MA
1
NU
Eikermann, MD, PhD,5 and Nicola Latronico,MD1,4.
Hospital, Brescia, Italy 2
Department of Physical Medicine and Rehabilitation, Spedali Civili University Hospital,
University of Brescia, Italy
Department of Anesthesia, Critical Care, and Pain Medicine, Massachusetts General Hospital
AC
5
Department of Medical and Surgical Specialties, Radiological Sciences and Public Health,
CE
4
National Heart and Lung Institute, Imperial College London, London, England
PT
3
ED
Brescia, Italy
and Harvard Medical School, Boston, Massachusetts, USA
Corresponding author Prof. Nicola Latronico Department of Anesthesia, Critical Care Medicine and Emergency Spedali Civili University Hospital Piazzale Ospedali Civili, 1 – 25123 Brescia, Italy Tel. +39-030 3995 561 (secretary)/764 (ICU) Fax +39-030 3995 779 Email [email protected]
1
ACCEPTED MANUSCRIPT
ABSTRACT
RI P
T
Purpose
We validated the Italian version of Surgical Optimal Mobility Score (SOMS) and evaluated its
NU
mortality in a mixed population of ICU patients.
SC
ability to predict intensive care unit (ICU) and hospital length of stay (LOS), and hospital
MA
Materials and Methods
We applied the Italian version of SOMS in a consecutive series of prospectively enrolled, adult
ED
ICU patients. SOMS level was assessed by ICU nurses twice a day, and twice a week by an expert mobility team (EMT). Zero-truncated Poisson regression was used to identify predictors
PT
for ICU and hospital LOS, and logistic regression for hospital mortality. All models were
AC
Results
CE
adjusted for potential confounders.
Of 98 patients recruited, 19 (19.4%) died in hospital, of whom 17 without and 2 with improved mobility level achieved during the ICU. SOMS improvement was independently associated with lower hospital mortality (OR: 0.07; 95% CI: 0.01-0.42) but increased hospital LOS (OR: 1.21; 95% CI: 1.10-1.33). A higher first morning SOMS on ICU admission, indicating better mobility, was associated with lower ICU and hospital LOS (rate ratio: 0.89, 95% CI: 0.80-0.99, and 0.84, 95% CI: 0.79-0.89, respectively).
2
ACCEPTED MANUSCRIPT
Conclusions
T
The first morning SOMS on ICU admission predicted ICU and hospital LOS in a mixed
RI P
population of ICU patients. SOMS improvement was associated with reduced hospital mortality
SC
but increased hospital LOS suggesting the need of optimizing hospital trajectories after ICU
MA
NU
discharge.
ED
Word count. Abstract: 224 words. Body text: 3,628ß. Number of figures: 2. Number of
PT
supplemental figures: 1. Number of tables: 6. Number of references: 35.
AC
neurologic patients.
CE
Key words: early mobilization; rehabilitation; safety; mortality; validation; surgical patients;
3
ACCEPTED MANUSCRIPT
Introduction
RI P
T
Early mobilization during the intensive care unit (ICU) stay has been advocated to mitigate the effects of muscle weakness[1-5]. Early rehabilitation includes a spectrum of
SC
interventions ranging from passive motion exercises to ambulation and to the use of novel technologies such as cycle-ergometry and transcutaneous electrical muscle stimulation[6].
NU
Although some patients are unable to progress through all the different steps of mobilization,
MA
even minimal motor activities can reduce muscle weakness and wasting. The beneficial effects of early mobilization in the medical ICU have been reported in
ED
several studies[1, 2]. Although not free of risks[3, 5, 7], its safety profile is reported to be good
PT
with low rates of complications even in patients who are traditionally not mobilized, such as those with femoral vein or artery catheters [8-10]. Different strategies have been proposed to
CE
modify potential barriers to early mobilization, such as changing vascular catheter location,
AC
careful scheduling of procedures, and improved sedation management [11, 12]. The Surgical Optimal Mobilization Score (SOMS), a five-point numerical rating scale to guide goal-directed early mobilization therapy, has been demonstrated to be a predictor of mortality as well as ICU and hospital length of stay (LOS) in a surgical ICU population[13]. In their original study, Kasotakis and colleagues studied the reliability of SOMS score in 113 functionally independent surgical ICU patients by comparing SOMS assigned by with those assigned by an expert mobility team, and they found an excellent agreement between the two.
4
ACCEPTED MANUSCRIPT
The authors also found that higher SOMS scores, indicating better mobility, were associated with
T
lower mortality and hospital and ICU LOS[13].
RI P
In this prospective cohort study, we assessed if SOMS can predict ICU and hospital LOS
SC
and hospital mortality in an Italian population of adult medical, surgical and neurologic ICU patients. We also evaluated the inter-rater reliability of the Italian version of SOMS and the
NU
safety of SOMS-guided mobilization in this mixed ICU population.
MA
Materials and Methods
The SOMS is an algorithm for goal-directed early mobilization in the ICU. It contains a
ED
numerical rating scale from 0 to 4 to quantify the patient's mobilization capacity. SOMS 0
PT
indicates that no mobilization should be considered since deemed to be futile, as for patients in terminal unstable clinical condition such as those with intracranial hypertension or severe
CE
systemic hemodynamic and respiratory insufficiency. SOMS 1 indicates that the patient can
AC
receive passive range of motion exercise while in bed and SOMS 2 that the patient can be sitting up in bed. SOMS 3 indicates that the patient is able to stand with or without assistance, and SOMS 4 is assigned to patients able to ambulate[13]. Development of the Italian version of the SOMS As a first step to use SOMS in our ICU, we provided an Italian translation of the original English version following the recommendations for a comprehensive multistep process for translating, adapting and cross-validating instruments (supplemental figure 1)[14]. Two qualified medical
5
ACCEPTED MANUSCRIPT
doctors whose native language was Italian, fluent in English and with knowledge in early
T
rehabilitation in the ICU, independently translated into Italian the original version of the SOMS
RI P
score, the associated instructions, and the drawings’ text. Thereafter, a consensus meeting was held to agree on a fully comprehensible and accurate Italian translation consistent with the
SC
original English text. The draft was back translated into English and compared with the original
NU
to develop the final Italian translation.
MA
Study design and setting
The study was a prospective observational study conducted at the general and neurologic ICU of
ED
the Department of Anesthesia, Critical Care and Emergency of the Spedali Civili of Brescia, a large regional university-affiliated hospital. The ICU has 10 beds, 6 general and 4 neurologic.
PT
The daytime staffing of the unit consists of one medical coordinator, one attending physician,
CE
two residents (4th and 5th-year residents of the School of Specialty in Anesthesia and Critical Care Medicine), and six critical care nurses. The night shift team consists of one attending
AC
physician, one resident, and four nurses. Physical therapists are not dedicated exclusively to the ICU, and provide general and respiratory physical therapy for 5 days a week based on a physiatrist-activated written protocol. The study was approved by the local Ethics Committee (Comitato Etico Provinciale di Brescia, June 17, 2013, Ref. N. 1383). Detailed written information was provided to the patients and family members about the study, and written informed consent to participate to the study was obtained from the patient whenever possible. In case of altered consciousness, the Ethics
6
ACCEPTED MANUSCRIPT
Committee waived the requirement for consent, as in Italy relatives are not regarded as legal
T
representatives of the patient in the absence of a formal designation. Written informed consent
RI P
was subsequently requested from all surviving patients as soon as they regained their mental competency. The study was conducted according to the principles expressed in the Declaration
SC
of Helsinki.
NU
Patients were eligible if they were 18 years or older and were expected to stay in the ICU for at least 72 hours. Patients were defined as medical, surgical or neurological (patient's class)
MA
according to the reason for ICU admission. Patients with predicted ICU stay of less than 72 hours, admitted for post-surgery monitoring or with unstable spine or in terminal condition were
PT
Patient management
ED
excluded.
CE
All patients were managed following the early-goal directed therapy guidelines [15] and a goal-
AC
directed sedation protocol aimed at minimizing the use of sedatives through daily interruption[16]. In non-neurologic patients, daily interruption of sedation was part of the “Awakening and Breathing Coordination, Delirium monitoring/management, and Early exercise/mobility” (ABCDE) bundle. SOMS assessment of the level of mobility anticipated to be accomplished during the morning shift was performed at fixed timed (see Study procedure) and not necessarily after awakening. However, the morning shift nurse integrated the entire information obtained from the night shift nurse with the actual patient’s condition in order to assign SOMS scores on a solid clinical base [17].
7
ACCEPTED MANUSCRIPT
Critically ill neurologic patients had continuous monitoring of cerebral hemodynamics, including
T
intracranial pressure, cerebral perfusion pressure and cerebro-vascular autoregulation
RI P
monitoring, according to pre-defined protocol [18]. For multimodal data acquisition, we used the Intensive Care Monitoring software system (ICM+, University of Cambridge, UK) running on
SC
bedside laptop computers[18]. Neurological severity was graded according to the Glasgow Coma
NU
Scale [19] and the Full Outline of UnResponsiveness scale [20]. Continuous sedation and analgesia with propofol and fentanyl was gradually reduced and then interrupted if the
MA
neurological condition, the systemic and cerebral hemodynamics and brain CT findings stabilized according to the neurointensivist in charge. Initial repeated interruptions of sedation up
ED
to 3 times a day were followed by definitive interruption if the patient could tolerate it with no
PT
excessive stress reactions. This practice of early interruption of sedation has been in place for
Staff training
CE
several years in our unit, and has been demonstrated to be safe also by other centers[21].
AC
From April to June 2013 all ICU nurses were trained by study members to apply the SOMS in simulated cases and during routine care. The ICU staff was also provided with written instructions for proper SOMS scoring, and they were involved in 3 educational meetings devoted to present the potential benefits of early mobilization in the ICU. From July 2013 to October 2013 we prospectively recruited a consecutive series of critically ill patients admitted to our ICU.
8
ACCEPTED MANUSCRIPT
Study procedures
T
Study investigators screened all ICU admissions daily in the morning to identify those patients
RI P
fulfilling the inclusion criteria. The early mobilization program was started on the day after
SC
enrollment.
The SOMS level was assessed twice a day for each patient by two nurses during the morning
NU
shift at least 30 minutes after handover from the night shift nurse (morning SOMS) and then
MA
after the lunch break (afternoon SOMS). The two nurses' evaluations were separated from one another by a maximum of 30 minutes to minimize the effect of clinical fluctuation. The morning
ED
SOMS level was defined as the level of mobility anticipated to be accomplished during the morning shift. The level of mobilization effectively reached was defined as achieved SOMS. All
PT
nurses performed the assessments independently, and were blinded to the other’s assessment
CE
both in the morning and afternoon sessions. In addition, the nurses also recorded any barrier or complication (see next section) related to mobilization[22]. An expert mobility team (EMT)
AC
including an intensivist and a rehabilitation physician assessed the patients twice a week on Tuesday and Thursday morning to predict mobilization capacity using their expertise and the SOMS algorithm. The team, which was blinded to the morning SOMS predicted by the nurses, assigned a new independent SOMS score (EMT-SOMS). Progress in mobility goal (delta SOMS) for every day followed the SOMS algorithm (figure 1). A delta SOMS of 1 or 0 indicated progress or, respectively, lack thereof in mobility goals. Dedicated study members
9
ACCEPTED MANUSCRIPT
collected all SOMS scores and recorded the achieved SOMS, as well as other relevant
T
information.
RI P
Safety monitoring of SOMS-guided mobilization
SC
To ensure the safety of the subjects, each day the nursing staff monitored any mobilization-
NU
associated adverse event (AE) that occurred during or within 30 minutes after mobilization. We classified as serious adverse events (SAE),any unintended injury that was due at least in part
MA
to incorrect mobilization and that exposed the patient to an increased risk of death, measurable disability or intervention to prevent permanent impairment or damage[24]. Since most SAE are
ED
preceded by clinically observable warning signs[24], we defined the following medical events to
PT
withheld mobility interventions: a decline in hemodynamic (mean arterial pressure < 65 mmHg or systolic blood pressure < 90 mmHg) or respiratory status (SpO2 <88%); hypertensive crisis
CE
(systolic blood pressure >180 mmHg); need for administration of a new vasopressor agent; need
AC
to increase the positive end-expiratory pressure on the ventilator or a change from spontaneous breathing or continuous positive airway pressure to assist control mode once in a weaning mode; symptoms of acute myocardial ischemia, or cardiac arrhythmias requiring anti-arrhythmic agent; accidental removal of a device (airways and cardiac devices, chest tubes, vascular access, wound or dressing); severe pain (numeric rating scale, NRS, of 4 or more in collaborative patients and need for analgesics in non-collaborative patients); fall, defined as an unplanned descent of the patient to the floor, and classified as with or without injury according to American Nurses
10
ACCEPTED MANUSCRIPT
Association, National Database of Nursing Quality Indicators [25]. Reported AEs and associated
T
complications were prospectively recorded by dedicated study members.
RI P
Data presentation and statistical analysis
SC
We expressed continuous variables as means (SD) or as medians (interquartile range, IQR) as appropriate. Discrete variables, including AE and other safety variables, were expressed as
NU
counts (percentage).
MA
We evaluated if SOMS can predict ICU and hospital LOS hospital and hospital mortality in order to replicate the results of the original SOMS study [13] in our mixed ICU population.
ED
Based on the original study [13], we anticipated an inverse correlation between the first morning
PT
SOMS and the ICU LOS of r ≥ 0.35. We calculated that a sample size of 98 patients would provide 80% power to detect such an association between the first morning SOMS and the ICU
CE
LOS with a two-sided α of 0.05.
AC
For ICU and hospital LOS, we used a zero-truncated Poisson regression model with rate ratios to identify predictors predictors. The following predicting variables were considered: age, first morning SOMS after ICU admission, delta SOMS, SAPS II score, comorbidity index, patient’s class (neurological versus others)[26], and use of vasopressors and renal replacement therapy. For the adjusted analysis, variables were included in the model if they were statistically significant in unadjusted analysis or if they were clinically relevant.
11
ACCEPTED MANUSCRIPT
For hospital mortality, potential predictors were identified using unadjusted logistic regression
T
with odds ratios (OR). As for LOS, for the adjusted analysis variables were included in the
RI P
model if they were statistically significant in unadjusted analysis or were clinically relevant. Overall model evaluation against the null model was obtained using the likelihood ratio test.
SC
Model calibration was assessed using the Hosmer-Lemeshow test, while the Area Under
NU
Receiver Operating Characteristic curve (AUROC) was used to assess discrimination. AUROC value >0.70, >0.80 and >0.90 were considered as acceptable, excellent and outstanding,
MA
respectively [27]. For each outcome we also estimated the variance explained by the model, R2. To assess SOMS inter-rater reliability in pairwise comparisons, we compared: a) SOMS
ED
assigned by the two nurses in the morning (morning SOMS), and b) in the afternoon (afternoon
PT
SOMS), and c) morning SOMS by the two nurses (the median value) versus EMT-SOMS. Interrater reliability was investigated using Spearman rho correlation coefficients with corresponding
CE
p-values among the three raters, and the percentage agreement. Correlation coefficients below
AC
0.40 were defined poor, between 0.41to 0.60 moderate, between 0.61 to 0.80 good, and above 0.80 very good [28].
For all analyses, a p-value of less than 0.05 was considered to indicate statistical significance. Statistical analyses were performed using STATA 13 (Stata Corp LP, College Station, Texas, USA).
12
ACCEPTED MANUSCRIPT
Results
RI P
T
During the study period, 187 patients were admitted to the ICU, 89 were excluded (41 not critically ill and admitted for postoperative monitoring only; 32 with an ICU stay of less than 72
SC
hours; 9 younger than 18 years; 7 in terminal condition), and 98 were recruited in the study. Demographic characteristics, admission diagnoses, patient severity and outcome are presented in
NU
table 1.
MA
In total, we obtained 854 morning SOMS evaluations: SOMS 0 was assigned in 41 cases (4.8%), SOMS 1 in 500 (58.5%), SOMS 2 in 230 (26.9%), SOMS 3 in 65 (7.6%) and SOMS 4 in 18
ED
(2.1%) cases.
PT
In 61 occasions (7.1%), the achieved SOMS was lower than morning SOMS, due to: need for sedation (n=24 evaluations out of 854), agitation (n=13), presence of femoral catheter (n=11),
CE
surgery (n=5), arterial hypotension (n=3), absence of resources for mobilization (n=3), patient
AC
refusal (n=3), delirium (n=3) and mild oxygen desaturation (n=1). There were 10 AEs (1.2% of all mobilization sessions), of which 1 (0.1%) was a SAE of hypotension requiring a vasopressor agent in a 65-year old patient with abdominal septic shock and a history of cardiac arrhythmias. This patient improved progressively and was safely discharged to a rehabilitation center, but died upon initiation of rehabilitation due to fatal ventricular dysrhythmia. Of the other 9 cases (4 of mild hypotension, 3 of mild pain and 2 falls from chair without injury), only 2 (0.2%) occurred in neurologic patients (2 falls).
13
ACCEPTED MANUSCRIPT
The median ICU LOS was 8 days (IQR= 4.0-13.5 days). In the unadjusted analyses, ICU LOS
T
was inversely associated with the first morning SOMS and delta SOMS (the higher the first
RI P
morning SOMS and delta SOMS the lower the ICU LOS), and directly associated with SAPS II scores and comorbidity index (table 2). In the adjusted analysis, ICU LOS was inversely
SC
associated with first morning SOMS (an improvement of 1 point of the SOMS score decreased
NU
the ICU LOS by 11.1%) and directly associated with SAPS II and comorbidity index (table 3). Hospital LOS had a median of 20 days (IQR= 12.0-33.5 days). In the unadjusted analyses it was
MA
inversely associated with first morning SOMS and patient's class (hospital LOS was lower in non-neurologic patients), and directly associated with delta SOMS (patients with improved
ED
mobility level during the ICU stay had longer hospital LOS), SAPS II scores and comorbidity
PT
index (table 2). In the adjusted analysis, hospital LOS was inversely associated with first morning SOMS (an improvement of 1 point of the SOMS score decreased hospital LOS by 16%)
CE
and patient's class, and directly associated with delta SOMS (an improved mobility level
AC
increased the hospital LOS by 20.7%) (table 3). Of 98 patients, 19 (19.4%) died in hospital, of whom 17 without and 2 with a progression in mobility goals. In the unadjusted analyses the hospital mortality was inversely associated with delta SOMS and directly associated with SAPS II (table 4). Both remained significant in the adjusted analysis (table 5). The fit of the multivariable model was good (Hosmer-Lemeshow test p-value: 0.97), and its overall accuracy in predicting mortality was excellent, with an AUROC of 0.89 (figure 2).
14
ACCEPTED MANUSCRIPT
Spearman rho correlation coefficients were very good for all comparisons, and the percentage
RI P
T
agreement between SOMS raters was always higher than 80% (table 6).
SC
Discussion
NU
We translated the original English version of SOMS in Italian using a rigorous methodology and
MA
we applied it in a mixed population of unselected medical, surgical and neurologic ICU patients. SOMS showed reliably when used in routine clinical practice to guide early mobilization. SOMS
ED
scores were applied by ICU staff with excellent inter-rater reliability, comparable to that reported by the developers of the scale. Reliability was excellent both in nurse-to-nurse and nurse-to-
PT
expert comparisons. The SOMS level of mobility predicted by nurses on the first morning after
CE
ICU admission was independently associated with ICU and hospital LOS. Improvement in the level of mobility achieved during the ICU stay was independently associated with reduced
AC
hospital mortality, but also with prolonged hospital LOS. These data support the hypothesis that SOMS has construct validity and that is a valid algorithm to select the optimal patient's mobility level. Moreover, they confirm the results of the original study showing that SOMS at ICU admission predicted ICU and hospital LOS[13]. SOMS-standardized mobilization was safely implemented by ICU nurses after an adequate training with application of SOMS in simulated and real cases. Absence of sufficient resources was the cause of failed mobilization in only 3 of 854 evaluations, indicating that SOMS can be
15
ACCEPTED MANUSCRIPT
implemented in routine ICU care with no supplemental cost or need for specialized equipment.
T
AE occurred in 1.2 percent of all mobilization sessions, which compares well with other studies
RI P
where figures varied from 0% to 22% [9]. The AE rate was excellent also in acutely ill neurologic patients, of whom one third could sit up in bed, stand or ambulate during the ICU
SC
stay, indicating that early mobilization can be safely implemented in this category of patients,
NU
provided that cerebral and systemic physiology are stabilized. Indeed, the neurocritical care can be an ideal setting for this intervention because immobility is common consequence of
MA
neurological dysfunction. Titsworth and colleagues reported similar findings in a population of 93 neurocritical care patients mobilized according to an 11-step Progressive Upright Mobility
ED
Protocol (PUMP) Plus algorithm [32]; however, PUMP Plus has not been validated and requires
PT
the acquisition of specialized equipment and dedicated personnel. Patients with pre-existing physical dysfunction may benefit less from early mobilization, but unlike the original study, we
CE
did not exclude patients with reduced functional independence (Barthel index score 70) before
AC
ICU admission. This may increase the generalizability of our results, further reinforcing that SOMS can predict LOS and mortality in unselected ICU patient populations. ICU LOS was reduced in patients with higher SOMS scores on the first morning after ICU admission and in those with less severe disease and burden of comorbid disease. Morning SOMS as a mobility level predicted to be accomplished during the morning shift can be considered itself a descriptor of patient’s severity. Clinician estimates of patient’s outcome in the ICU are often accurate, reflecting a great deal of experience [33]. This finding is in line with the result of the original study showing that SOMS predicts ICU LOS as well as the APACHE score [13]. More
16
ACCEPTED MANUSCRIPT
than 60 percent of our patients were acutely ill neurologic patients, and hence, our results also
T
confirm those of Titsworth showing a reduced duration of the ICU stay upon implementation of
RI P
an early mobility program among neurocritical care unit patients [32].
SC
As for ICU LOS, hospital LOS was reduced in patients with higher first morning SOMS, in those with less severe disease and burden of comorbid disease, and in those with a surgical or medical
NU
cause of ICU admission compared to neurologic patients. Unexpectedly, hospital LOS was increased in patients with improvement of the mobility level during the ICU stay. Reasons for
MA
this result remain speculative because we did not plan specific follow-up investigations after ICU discharge. In the original study enrolling a homogeneous cohort of surgical ICU patients, a
ED
higher SOMS score was associated with reduced hospital stay, but delta SOMS was not
PT
evaluated[13]. Delta SOMS has a completely different meaning compared to the first morning SOMS, representing the improvement in mobility goals achieved during the ICU stay and not a
CE
proxy of disease's severity. We speculate that transfer of ICU patients to several different wards,
AC
each with specific discharge policy, may have influenced the results of our study. Future studies should consider the optimization of post-ICU trajectories within the acute-care hospital in order to properly evaluate the multidimensional aspects influencing the hospital LOS. This study is the first to demonstrate an independent association between improvement of mobility during the ICU stay and reduced mortality. Other studies have demonstrated improved functional independence at hospital discharge with a strategy of whole-body rehabilitation compared to standard care, but mortality was unchanged [1]. Although promising, this result
17
ACCEPTED MANUSCRIPT
must be interpreted with caution. Due to the observational nature and the small sample size of
T
the study, we could only consider a limited number of covariates such as the disease severity and
RI P
comorbidity in our multivariable model, while mortality in critically ill patients is influenced by multiple complex factors. Predictive modeling is difficult when considering the numerous
SC
clinical elements that occur after critical illness and their interplay[34, 35].
NU
Future efficacy trials with sufficient power and longer-term outcomes are needed to establish if SOMS-guided early mobilization may reduce mortality and functional disability in survivors of
ED
MA
critical illness.
PT
Conclusion
CE
Use of the SOMS algorithm to guide optimal patient's mobility level was safely implemented by ICU nurses in the routine care of a mixed population of medical, surgical and neurologic ICU
AC
patients after adequate training.
Inter-rater reliability of SOMS scores was excellent both in nurse-to-nurse and nurse-to-expert mobility team comparisons. Absence of sufficient resources was a rare cause of failed mobilization, indicating that SOMS can be implemented in routine ICU care with no supplemental cost or need for specialized equipment.
18
ACCEPTED MANUSCRIPT
Improvement in the level of mobility achieved during the ICU stay was independently associated
RI P
T
with reduced ICU length of stay and hospital mortality.
SC
Abbreviations
NU
AEs, adverse events during the ICU stay.
MA
EMT, expert mobility team.
LOS, length of stay.
PT
SAEs, serious adverse events.
ED
ICU, intensive care unit.
CE
SAPS II, Simplified Acute Physiology Score II.
AC
SOMS, Surgical Optimal Mobility Score.
Competing interests The authors have no competing interests to declare.
19
ACCEPTED MANUSCRIPT
Authors’ contributions
S. Piva, C. Minelli and N. Latronico drafted the manuscript.
RI P
T
S. Piva, M. Eikermann and N. Latronico were responsible for designing the study.
SC
S. Piva, N. Latronico, I. Moreno-Duarte, M. Eikermann and C. Minelli were responsible for data
NU
analysis and contributed to the revision of the manuscript.
simulated cases and during routine care.
MA
S. Piva, G. Dora and C. Sottini were responsible for training the ICU staff in using SOMS in
ED
S. Piva and G. Dora were responsible for organizing the expert mobility team. F. Turla and S. Mazza were responsible for providing continuous support to the critical care
PT
nurses in assigning SOMS score.
CE
S. Piva, F. Turla, S. Mazza provided written instructions for proper SOMS scoring, and
AC
organized the educational meetings on early ICU mobilization. S. Piva and N. Latronico translated the English version of SOMS in Italian and drafted the initial version of the paper.
M. Michelini and P. D'Ottavi were responsible for collecting and recording the SOMS scores, the achieved level of mobility, the clinical and demographic data and follow-up, and contributed to data analysis and interpretation. All authors read and approved the final version of the manuscript.
20
ACCEPTED MANUSCRIPT
1.
RI P
T
References
Schweickert WD, Pohlman MC, Pohlman AS, Nigos C, Pawlik AJ, Esbrook CL, et al.
SC
Early physical and occupational therapy in mechanically ventilated, critically ill patients: a randomised controlled trial. Lancet 2009; 373:1874-82
Burtin C, Clerckx B, Robbeets C, Ferdinande P, Langer D, Troosters T, et al. Early
NU
2.
MA
exercise in critically ill patients enhances short-term functional recovery. Critical care medicine 2009; 37:2499-505
Kayambu G, Boots R, Paratz J. Physical therapy for the critically ill in the ICU: a
ED
3.
systematic review and meta-analysis. Crit Care Med 2013; 41:1543-54 Eikermann M, Latronico N. What is new in prevention of muscle weakness in critically
PT
4.
5.
CE
ill patients? Intensive Care Med 2013; 39:2200-3 Parker A, Sricharoenchai T, Needham DM. Early Rehabilitation in the Intensive Care
AC
Unit: Preventing Physical and Mental Health Impairments. Current physical medicine and rehabilitation reports 2013; 1:307-14 6.
Berney S, Haines K, Denehy L. Physiotherapy in critical care in australia.
Cardiopulmonary physical therapy journal 2012; 23:19-25 7.
Needham DM, Davidson J, Cohen H, Hopkins RO, Weinert C, Wunsch H, et al.
Improving long-term outcomes after discharge from intensive care unit: report from a stakeholders' conference. Crit Care Med 2012; 40:502-9
21
ACCEPTED MANUSCRIPT
8.
Perme C, Nalty T, Winkelman C, Kenji Nawa R, Masud F. Safety and Efficacy of
9.
RI P
Observational Study. Cardiopulm Phys Ther J 2013; 24:12-7
T
Mobility Interventions in Patients with Femoral Catheters in the ICU: A Prospective
Sricharoenchai T, Parker AM, Zanni JM, Nelliot A, Dinglas VD, Needham DM. Safety
SC
of physical therapy interventions in critically ill patients: a single-center prospective evaluation
10.
NU
of 1110 intensive care unit admissions. J Crit Care 2014; 29:395-400 Fan E, Cheek F, Chlan L, Gosselink R, Hart N, Herridge MS, et al. An official American
MA
Thoracic Society Clinical Practice guideline: the diagnosis of intensive care unit-acquired weakness in adults. Am J Respir Crit Care Med 2014; 190:1437-46 Needham DM, Korupolu R, Zanni JM, Pradhan P, Colantuoni E, Palmer JB, et al. Early
ED
11.
PT
physical medicine and rehabilitation for patients with acute respiratory failure: a quality improvement project. Archives of physical medicine and rehabilitation 2010; 91:536-42 Engel HJ, Tatebe S, Alonzo PB, Mustille RL, Rivera MJ. Physical therapist-established
CE
12.
AC
intensive care unit early mobilization program: quality improvement project for critical care at the University of California San Francisco Medical Center. Physical therapy 2013; 93:975-85 13.
Kasotakis G, Schmidt U, Perry D, Grosse-Sundrup M, Benjamin J, Ryan C, et al. The
surgical intensive care unit optimal mobility score predicts mortality and length of stay. Critical care medicine 2012; 40:1122-8 14.
Wild D, Grove A, Martin M, Eremenco S, McElroy S, Verjee-Lorenz A, et al. Principles
of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes (PRO) Measures: report of the ISPOR Task Force for Translation and Cultural
22
ACCEPTED MANUSCRIPT
Adaptation. Value in health : the journal of the International Society for Pharmacoeconomics and
Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving
RI P
15.
T
Outcomes Research 2005; 8:94-104
Sepsis Campaign: international guidelines for management of severe sepsis and septic shock,
Girard TD, Kress JP, Fuchs BD, Thomason JW, Schweickert WD, Pun BT, et al.
NU
16.
SC
2012. Intensive Care Med 2013; 39:165-228
Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically
MA
ventilated patients in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial. Lancet 2008; 371:126-34
Vasilevskis EE, Ely EW, Speroff T, Pun BT, Boehm L, Dittus RS. Reducing iatrogenic
ED
17.
PT
risks: ICU-acquired delirium and weakness--crossing the quality chasm. Chest 2010; 138:122433
Rasulo FA, Girardini A, Lavinio A, De Peri E, Stefini R, Cenzato M, et al. Are optimal
CE
18.
AC
cerebral perfusion pressure and cerebrovascular autoregulation related to long-term outcome in patients with aneurysmal subarachnoid hemorrhage? J Neurosurg Anesthesiol 2012; 24:3-8 19.
Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A practical
scale. Lancet 1974; 2:81-4 20.
Wijdicks EF, Bamlet WR, Maramattom BV, Manno EM, McClelland RL. Validation of a
new coma scale: The FOUR score. Ann Neurol 2005; 58:585-93
23
ACCEPTED MANUSCRIPT
21.
Skoglund K, Hillered L, Purins K, Tsitsopoulos PP, Flygt J, Engquist H, et al. The
22.
RI P
with severe traumatic brain injury. Neurocrit Care 2014; 20:413-26
T
neurological wake-up test does not alter cerebral energy metabolism and oxygenation in patients
Leditschke IA, Green M, Irvine J, Bissett B, Mitchell IA. What are the barriers to
Brennan TA, Localio AR, Leape LL, Laird NM, Peterson L, Hiatt HH, et al.
NU
23.
SC
mobilizing intensive care patients? Cardiopulmonary physical therapy journal 2012; 23:26-9
Identification of adverse events occurring during hospitalization. A cross-sectional study of
MA
litigation, quality assurance, and medical records at two teaching hospitals. Ann Intern Med 1990; 112:221-6
Jones DA, DeVita MA, Bellomo R. Rapid-response teams. N Engl J Med 2011; 365:139-
ED
24.
PT
46
Currie LM. Fall and injury prevention. Annu Rev Nurs Res 2006; 24:39-74
26.
Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying
CE
25.
40:373-83
AC
prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;
27.
David W. Hosmer, Jr. Applied Logistic Regression, 3rd Edition.
28.
David W. Hosmer J, Stanley Lemeshow, Rodney X. Sturdivant: Applied Logistic
Regression, Third Edition, 2013 29.
Morris PE, Goad A, Thompson C, Taylor K, Harry B, Passmore L, et al. Early intensive
care unit mobility therapy in the treatment of acute respiratory failure. Crit Care Med 2008; 36:2238-43
24
ACCEPTED MANUSCRIPT
30.
Elliott D, McKinley S, Alison J, Aitken LM, King M, Leslie GD, et al. Health-related
T
quality of life and physical recovery after a critical illness: a multi-centre randomised controlled
31.
RI P
trial of a home-based physical rehabilitation program. Crit Care 2011; 15:R142 Skinner EH, Berney S, Warrillow S, Denehy L. Development of a physical function
SC
outcome measure (PFIT) and a pilot exercise training protocol for use in intensive care. Critical
NU
care and resuscitation : journal of the Australasian Academy of Critical Care Medicine 2009; 11:110-5
Titsworth WL, Hester J, Correia T, Reed R, Guin P, Archibald L, et al. The effect of
MA
32.
increased mobility on morbidity in the neurointensive care unit. Journal of neurosurgery 2012;
Rocker G, Cook D, Sjokvist P, Weaver B, Finfer S, McDonald E, et al. Clinician
PT
33.
ED
116:1379-88
predictions of intensive care unit mortality. Crit Care Med 2004; 32:1149-54 Latronico N. Prediction is very difficult, especially about the future*. Crit Care Med
35.
AC
2015; 43:505-6
CE
34.
Latronico N, Herridge MS. Unraveling the myriad contributors to persistent diminished
exercise capacity after critical illness. Intensive care medicine 2015;
25
ACCEPTED MANUSCRIPT
Figure 1. Surgical Optimal Mobility Score.
SOMS 0: No activity
SOMS 1: PROM, upright in bed
SOMS 2: Sitting up
SOMS 3: Standing
SOMS 4: Ambulating
MA
NU
SC
1. Stable spine, no SCI. 2. ICP < 20 mmHg 3. Not a moribund patient 1. Follow simple commands. 2. No open spinal drains, EVD, peritoneum, chest. 3. No femoral CVVH lines.
RI P
Reproduced with permission of the Editor Lippincott Williams & Wilkins.
T
CVVH, continuous veno-venous hemofiltration; EVD, extracranial ventricular drains; ICP, intracranial pressure; PROM, passive range of motion; SCI, spinal cord injury.
1. Stand twice with minimal assist. 2. Steps in place with minimal assist.
AC
CE
PT
ED
1. Bilateral quadriceps strength 3/5 or more 2. Sits without support 3. No weight -bearing restrictions.
26
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
Figure 2. Receiver-operating characteristic curve for hospital mortality, multivariable adjusted regression.
27
ACCEPTED MANUSCRIPT
Table 1. Demographic characteristics of participants, admission diagnoses and outcome. ICU, intensive care unit. SAPS II, Simplified Acute Physiology Score.
RI P
T
Measure 61.5 (16.4) 67 (68.3%)
PT
CE
Mechanical ventilation [number of patients (percentage)] SAPS II score [mean (SD)] Vasopressor therapy [number of patients (percentage)] Renal replacement therapy [number of patients (percentage)] ICU length of stay, days [mean (SD)] Neurologic patients Surgical patients Medical patients Hospital length of stay, days [mean (SD)] Neurologic patients Surgical patients Medical patients Hospital mortality [number (percentage)]
AC
60 (61.2) 18 15 7 3 2 2 13 15 (15.3) 4 2 4 5 23 (23.5) 11 8 4
SC NU
ED
MA
Variables Age, years [mean (SD)] Male [number (percentage)] Diagnosis at ICU admission [number (percentage)] Neurological diseases head trauma subarachnoid hemorrage intracerebral hematoma encephalitis seizure/epilepsy post-cardiac arrest encephalopathy other diagnoses Surgical diseases major vascular surgery abdominal surgery polytrauma other diagnoses Medical diseases sepsis/septic shock acute respiratory distress syndrome other diagnoses
76 (77.5%) 41.1 (15.7) 29 (29.5) 6 (6.1) 9.5 (6.4) 9.8 (5.7) 11.3 (8.6) 7.6 (6.5) 25.5 (23.7) 24.0 (26.4) 35.5 (17.7) 23 (17.4) 19 (19.3)
28
ACCEPTED MANUSCRIPT
Table 2. Predictors of the intensive care unit (ICU) and hospital length of stay, unadjusted analysis.
Variables
IRR
95% C.I.
p
SC
ICU length of stay
RI P
T
SOMS, surgical optimal mobility score. IRR, incidence risk ratio, is a hazard ratio and is interpreted in the same way. SAPS II, Simplified Acute Physiology Score. Patient's class indicates the cause of ICU admission (medical, surgical, neurological), with neurological patients as the reference class.
0.904
0.825-0.991
0.032
Delta SOMS
0.820
0.716-0.939
<0.004
Patient's class
1.072
0.9304-1.236
0.334
SAPS II score
1.015
1.010-1.019
<0.001
Comorbidity Index
1.212
MA
NU
First morning SOMS
1.129-1.301
<0.001
Hospital length of stay 0.869
Delta SOMS
1.086
Patient's class
0.855
SAPS II score Comorbidity Index
0.822-0.920
ED
First morning SOMS
<0.001 0.051
0.787-0.931
<0.001
1.009
1.007-1.012
<0.001
1.274
1.221-1.329
<0.001
AC
CE
PT
0.999-1.181
29
ACCEPTED MANUSCRIPT
Table 3. Predictors of the intensive care unit (ICU) and hospital length of stay (LOS), multivariable adjusted analysis.
RI P
T
SOMS, surgical optimal mobility score. IRR, incidence risk ratio, is a hazard ratio and is interpreted in the same way. SAPS II, Simplified Acute Physiology Score. Patient's class indicates the cause of ICU admission (medical, surgical, neurological), with neurological patients as the reference class.
Variable
SC
R2= 0.1048 (ICU LOS) and 0.0837 (hospital LOS).
IRR
95% C.I.
p
SAPS II
1.011
Comorbidity Index
1.112
Constant
5.983
Variable
IRR
0.802-0.986
0.025
1.006-1.016
<0.001
1.022-1.209
0.013
4.402-8.131
<0.001
95% C.I.
p
MA
0.889
ED
First morning SOMS
NU
ICU length of stay
Hospital length of stay
1.207
First morning SOMS Patient's class
Constant
0.840
0.789-0.893
<0.001
0.908
0.834-0.989
0.027
1.005
1.002-1.009
<0.001
1.250
1.190-1.313
<0.001
18.640
15.522-22.384
<0.001
AC
Comorbidity Index
<0.001
CE
SAPS II
1.100-1.325
PT
Delta SOMS
30
ACCEPTED MANUSCRIPT
Table 4. Predictors of hospital mortality, unadjusted regression analysis.
T
SOMS, surgical optimal mobility score. OR, odds ratio. SAPS II, Simplified Acute Physiology Score. CVVH, Continuous Venous-Venous Hemofiltration. Patient's class indicates the cause of ICU admission (medical, surgical, neurological), with neurological patients as the reference class.
95% C.I.
First morning SOMS
0.995
0.494-2.007
0.991
Delta SOMS
0.066
0.014-0.308
0.001
Patient's class
0.363
0.130-1.018
SAPS II score
1.077
1.032-1.124
0.001
Comorbidity Index
1.538
Vasopressors
1.306
CVVH
1.906
0.887-2.667
0.125
0.426-4.003
0.641
0.320-11.353
0.479
NU
0.054
AC
CE
PT
ED
MA
p
SC
RI P
OR
31
ACCEPTED MANUSCRIPT
Table 5. Predictors of hospital mortality, multivariable adjusted analysis. SOMS, surgical optimal mobility score. SAPS II, Simplified Acute Physiology Score. Patient's class indicates the cause of ICU admission, either medical, surgical or neurological. OR= Odds ratio. C.I.= Confidence Interval.
RI P
T
R2= 0.2861.
OR
95% C.I.
Delta SOMS
0.066
0.010-0.417
SAPS II
1.071
1.007-1.140
0.030
Constant
0.222
0.010-4.689
0.033
P 0.004
AC
CE
PT
ED
MA
NU
SC
Variable
32
ACCEPTED MANUSCRIPT
Table 6. Inter-rater reliability of Surgical Optimal Mobility Score (SOMS).
Spearman rho
P-value
0.982
<0.001
0.984
<0.001
80.32%
0.813
<0.001
PT
RI P
Percentage of agreement
T
EMT, expert mobility team. *We used the median morning SOMS assigned by two critical care nurses
Raters
94.39%
0.981
<0.001
Agreement between the afternoon SOMS assigned by two critical care nurses
98.55%
0.984
<0.001
84.40%
0.840
<0.001
SC
All patients 92.76%
NU
Agreement between the morning SOMS assigned by two critical care nurses
ED
Agreement between morning SOMS* and EMT-SOMS
98.30%
MA
Agreement between the afternoon SOMS assigned by two critical care nurses
Neurologic patients
AC
CE
Agreement between the morning SOMS assigned by two critical care nurses
Agreement between morning SOMS* and EMT-SOMS
33