Evidence-based nursing interventions and guidelines for prone positioning of adult, ventilated patients: a systematic review.
Although the prone positioning of a critically ill patient poses a challenge to nursing interventions, it remains the responsibility of nurses to develop a way to provide the same basic and intensive care to those patients lying prone as to those lying supine. The purpose of this study was firstly to conduct a systematic review of the literature as exploration and description of the evidence in support of the beneficial nursing interventions during prone positioning of ventilated patients, and secondly to develop evidence-based nursing guidelines for the nursing process.
This exploratory, descriptive and retrospective systematic review includes data from 45 clinical trials, with a total population of 2 148 patients. Data was extracted onto data abstraction forms, assessed for methodological quality and finally summarised in evidence tables. All statistical calculations for the meta-analysis were performed by the RevMan 4.2.8 program. Prone positioning showed significant (p < 0.0001) increases in the partial pressure of oxygen in arterial blood (Pa[O.sub.2]) weighted mean difference (WMD = 11.43) and the partial pressure of oxygen in arterial blood/fraction of inspired oxygen (Pa[O.sub.2]/Fi[O.sub.2]) ratio (WMD = 21.58, 95% CI = 11.36; 31.8). The effects of complications, oxygenation and haemodynamic outcomes compared with the different prone-positioning protocols produced inconclusive results. Nursing guidelines for prone positioning were developed based on the best available evidence. The lack of related articles on nursing care of prone positioning was a drawback. Based on these results, recommendations are made towards further study on the nursing care of prone-positioned patients.
Keywords: prone positioning; ventilated patients; intensive care; evidence-based; nursing guidelines
Hoewel maagleposisionering van die pasient, wat in 'n kritieke toestand is, 'n uitdaging aan verpleegsorg bied, is dit steeds die verantwoordelikheid van die verpleegkundige om dieselfde basiese en kritieke sorg aan pasiente te lewer in die maagleposisie as aan pasiente in die rugliggende posisie. Die doel van hierdie navorsing is eerstens om 'n sistematiese ontleding van die literatuur te doen as verkennende en beskrywende bewyse van die voordelige verpleegintervensies gedurende maagleposisionering van geventileerde pasiente, en tweedens om navorsingsgebaseerde verpleegriglyne op te stel wat die verpleegproses betref.
Hierdie verkennende, beskrywende en retrospektiewe sistematiese ontleding sluit data in van 45 kliniese proewe met 'n totale populasie van 2 148 pasiente. Data is deur middel van 'n gestandardiseerde data-ontledingsvorm verkry, vir die metodologiese kwaliteit daarvan geassesseer en in dataonttrekkingstabelle opgesom. Alle statistiese verwerkings vir die meta-analise is deur die RevMan 4.2.8-program gedoen. Maagleposisionering het beduidende toenames (p < 0.0001) in die arteriele suurstofdruk (Pa[O.sub.2]) geweegde gemiddelde verskil (GMV = 11.43) en die arteriele suurstofdruk/fraksie van geinspireerde suurstof-verhouding (Pa[O.sub.2]/Fi[O.sub.2]) (GMV = 21.58, 95% CI = 11.36; 31.8) getoon. Die effek van die komplikasie-, oksigenerings-, en hemodinamiese uitkomste in vergelyking met die verskillende maagleposisioneringsprotokolle het onbeduidende resultate gelewer. Verpleegriglyne vir maagleposisionering is opgestel volgens die beste beskikbare navorsingsbewyse. Die tekort aan verpleegsorgverwante artikels oor maagleposisionering word as 'n leemte beskou. Op grond van hierdie resultate word aanbevelings gemaak vir verdere studies oor die verpleegsorg van maagliggende pasiente.
Prone positioning improves oxygenation in patients with acute respiratory distress syndrome (ARDS) by shifting blood flow to the undamaged and better-ventilated regions of the lung (Gattinoni, Tognoni, Pesenti, Taccone, Mascheroni, Labarta, Malacrida, Giulio, Fumagalli, Pelosi, Brazzi & Latini, 2001:568). ARDS is associated with high mortality rates. According to the ARDS Foundation (n.d.), the extrapolated incidence of ARDS in South Africa was 24 512 out of an estimated population of 44 448 470 in 2004, Prone positioning challenges the normal day-to-day clinical practice of intensive care nurses. This procedure can be done with the help of a few experienced colleagues, but poses some risk to the patient. These risks include unplanned extubation, accidental removal of intravenous and arterial lines and underwater chest drain tubes, obstruction of invasive lines, pressure sores and joint contractures (McAuley, 2001:12).
Patients experience facial swelling, expel large amounts of oral and nasal secretions and may develop pressure sores on weight-bearing parts of the body (Balas, 2000:24). Unintended extubation, airway obstruction and difficulties coordinating the ventilator can result in critical events. To minimise these risks, it is important to prepare appropriately and anticipate potential complications. Most complications described in literature can be avoided by means of effective planning and nursing care of these patients.
According to Curley (1999:397), clinicians are currently attempting to design safe and effective guidelines for the use of prone positioning in critically ill patients with ARDS. Nurses should be able to assess, plan, implement and evaluate this procedure independently (Grossman & Bautista, 2002:34). Curley (1999:397) suggests that evidence-based protocols and efficacy studies for nurses should be designed to determine the care requirements of these patients. The purpose of this research was firstly to conduct a systematic review in order to explore and describe the evidence in support of the beneficial nursing interventions during prone positioning of ventilated patients, and secondly to develop evidence-based nursing guidelines for the nursing process.
The research design for this study can be described as an exploratory, descriptive and retrospective systematic review. Cook, Mulrow and Haynes (1997:376) describe a systematic review as a process that involves scientific strategies to synthesise the results of multiple primary studies that address a specific clinical question.
Criteria for inclusion and exclusion of studies
Berman and Parker (2002:11) promote inclusion and exclusion criteria in a meta-analysis to safeguard against selection bias. A systematic review was performed in this present study using experimental study designs (randomised controlled trials--RCTs) as well as comparative, non-randomised and observational studies as evidence. Selected studies included a population of adult or paediatric subjects that were ventilated and turned into the prone position. In addition, the search strategy was restricted to articles published or translated into English. According to Moher, Cook, Eastwood, Olkin, Rennie and Stroup (2000:1452), "[l]anguage-restricted meta-analyses [overestimate] the treatment effect by only 2% on average compared with language-inclusive meta-analyses". Studies that included neonates or animals were excluded.
Evidence was gathered using bibliographical databases of systematic reviews and articles that might provide definitive primary research data. A well-formulated question guides the entire review process (Stamm, Lawrence & Richardson, 1998:2). The question guiding this study was: Which nursing interventions during prone positioning will benefit the patient and reduce or eliminate complications? Glasziou, Irwig, Bain and Colditz (2001:17) suggest that a systematic approach be taken to break the study question down into components using a Venn diagram. The components nursing interventions, prone positioning and different outcomes--were searched. Studies were located using the following search strategies: The Cochrane Controlled Trials Register (CCTR), MEDLINE, ProQuest, EBSCOhost, CINAHL, the Internet, Citation Search and manual searching.
The data collection process comprised the systematic extraction of relevant data onto standardised data abstraction forms. Hedin and Kallestal (2004:28) suggest that data abstraction forms be created in order to objectively note data from the original articles. A data abstraction form was adapted from the Handbook for compilation of reviews on interventions in the field of public health (Hedin & Kallestal, 2004:28) to measure study characteristics and outcome measures. The data from this form was used in descriptive summaries and evidence tables. Data compiled in tables was then used for qualitative description and quantitative analysis (meta-analysis).
The evidence was grouped into the following categories: demographic data, ventilation strategies, prone protocols, responder and non-responder groups, mortality, prone complications, responder outcomes, overall oxygenation outcomes and haemodynamic outcomes. An additional table was created to group exclusions and contra-indications to prone positioning. According to Hammersley (2002:1), it is important that "the practical use of research [take] in the whole range of findings on a topic, not just the results from one or two studies. For this reason, reviews play a crucial role as a bridge between research and related areas of policymaking and practice". Systematic reviews are used to integrate existing information efficiently and provide data for rational decisionmaking (Mulrow, 1994:597).
Fifty-six articles were identified, 11 of which were excluded from this review, leaving a total of 45 articles used in the final review. Data was entered onto a spreadsheet and captured into the RevMan 4.2.8 software program. Evidence gained from RCTs was used for quantitative analysis, which included a comparison between the supine and the prone position. Outcomes were chosen based on the available outcomes measured in each randomised trial. A subgroup analysis was performed to compare the results to the different protocols of prone positioning and included the following: sub-group analysis by responders and non-responders; volume-controlled and pressure-controlled ventilation; prone position with support under the chest and pelvis or without support; time in prone position; number of turns from supine to prone position; the number of personnel involved in the process; and whether pronepositioned patients were both sedated and paralysed, or only sedated.
Assessment of methodological quality
A quality rating instrument was used to assess the methodological quality of each individual study, and was included with the data abstraction form. Each study was given a level of evidence in accordance with the Oxford Centre for Evidence-based Medicine levels of evidence and thereafter graded according to grades of recommendation (Phillips, Ball, Sackett, Badenoch, Straus, Haynes & Dawes, 2001:1). This resulted in studies that rated A to D, in which A can be associated with studies of good quality and D with those of poor quality. The study quality of RCTs was evaluated in more detail by using The Jadad Quality Assessment Instrument for Clinical Trials (Jadad, Moore, Carroll, Jenkinson, Reynolds, Gavaghan & McQuay, 1996:1). This instrument assesses randomisation, doubleblinding, withdrawals and dropouts according to a scoring system with a score from 0-5. RCTs with a Jadad score of more than or equal to 3 can be considered as a medium to good quality study. Studies with a Jadad score of less than 3 are considered to be poor in methodological quality, thereby reducing the validity of the systematic review.
Data synthesis was achieved by a narrative summary of studies and, where appropriate, by statistically combining the data produced by individual studies (Briggs, 2001:4). All statistical calculations were performed by RevMan available from the Cochrane Collaboration (Review Manager, 2003). Meta-analysis for this systematic review involved the calculation of an estimate of the chance variation from the selected studies, namely the confidence interval (CI). Evidence gained from RCTs was presented in a graph which shows a point estimate of each study's effect size, with the CI indicated by a horizontal line on either side of the point. This is also known as a forest plot (Hetherington, 2004:1). The treatment effect, effect size and the study size or variance were measured in each study.
An odds-ratio was used to compare dichotomous outcomes and the weighted mean difference (WMD) to measure continuous data. Statistical heterogeneity was tested using a chi-square statistic (to test whether the two variables were independent or related). The degrees of freedom were calculated to determine the significance of the value of the statistic (Burns & Grove, 2001:571). Where the heterogeneity shows significance (P-value > 0.1), a random-effects model was used. Even though a significant chi-square value indicates difference, the magnitude of the difference is not revealed by the analysis (Burns & Grove, 2001:572).
Characteristics of studies
Forty-five clinical trials involving prone positioning were identified, with a total population of 2 148 patients. Fortytwo clinical trials had adult populations and three had paediatric populations (prone positioning protocols were found to be similar in both adult and paediatric populations). There were 13 RCTs, only seven of which could be included for quantitative analysis. Thirty-two non-randomised controlled trials consisted of crossover designs, retrospective analysis or chart reviews, clinical follow-up studies, prospective clinical cohort studies, case-series trials, physiological studies and descriptive and observational studies. The methodological quality assessment showed 18 A-grade, 20 B-grade and 3 Cgrade studies according to the grades of recommendation. Eight RCTs had a Jadad score below 3, five had a Jadad score of 3 and no studies had a Jadad score above 3. Participants had a mean age of 49.7 years and a mean weight of 71 kg. The percentage of patients with primary ARDS was 59%, and 41% of patients in these studies presented with secondary ARDS.
All the haemodynamic parameters increased when the patient was turned from the supine to the prone position, but only pulmonary artery wedge pressure (PAWP) showed statistical significance. The following results were available from four studies with a total sample size of 60 subjects in each position. The test for overall effect, p = 0.0009, represents a significantly higher PAWP in the prone position. PAWP was between 13.4 and 24.4 mmHg in the prone position and between 10.5 and 22 mmHg in the supine position (WMD = 2.94 mmHg). The normal value is 2-12 mmHg (Beers & Berkow, 1999:1625). These results correlate to those of a recent study performed by Ferguson, Meade, Hallett and Stewart (2002:1073). In their study the mean maximum PAWP reading among patients was 22.5 mmHg (95% CI 21.2-23.8) and the mean median PAWP was 16.6 mmHg (95% CI 15.6-17.5).
Partial pressure of oxygen in arterial blood (Pa[O.sub.2]) and partial pressure of oxygen in arterial blood/fraction of inspired oxygen (Pa[O.sub.2]/Fi[O.sub.2]) ratio (see figure 1) showed statistically significant improvements from the supine to the prone positions. These results were expected and in agreement with the outcomes of other trials on prone positioning. The Pa[O.sub.2]/Fi[O.sub.2] ratio results are far below the normal range (80 to 100 mmHg), which confirms the diagnoses of ARDS (Thelan, Urden, Lough & Stacy, 1998:662).
Nineteen trials classified their patients into responders (if they met certain oxygenation improvement criteria) and non-responders (if they did not meet these criteria). The patients were divided into a responder and nonresponder ratio, with 25% of patients classified as nonresponders and 75% as responders to prone positioning.
[FIGURE 1 OMITTED]
Prone positioning can also be combined with other treatment modalities. Among the clinical trials included, 14% of studies included inhaled nitric oxide therapy and 13% included extracorporeal membrane oxygenation (ECMO), airway pressure release ventilation (APRV) with unsupported spontaneous breathing, continuous rotation and sigh breaths. Prone positioning was mostly (73% of all studies) used effectively as the only treatment modality.
In 28 trials patients were both sedated and paralysed and in eight trials, patients were sedated only. The heart rate increased by a WMD of 3.06 beats/minute in sedated and paralysed patients. The heart rate increased (insignificantly) more with sedation and paralysis than with sedation only. This may be due to the type of muscle relaxant used. Many of the studies included mentioned the use of Pancuronium bromide, a non-depolarising curaremimetic muscle relaxant which acts as an acetylcholine antagonist on neuromuscular junctions. A side effect of this drug is that it raises heart rate moderately (Wikipedia, n.d. Pancuronium). Pa[O.sub.2] increased significantly in patients who were sedated and paralysed.
Complications that occurred with prone positioning were found throughout the literature. These are listed in Table 1 from the most common complication to the least. Pressure sores were found on the face, thorax and anterior chest wall, shoulders, iliac crest, knees, lips, tongue, forehead, hips, chin, genitalia, upper chest wall (necrosis) and ears.
It was noted that endotracheal tube dislodgement and the loss of arterial or venous access occurred more often in patients in the supine than with those in the prone position. Most patients in the clinical trials included represented with mild cutaneous damage or facial oedema to some degree. Unfortunately these complications were not measured or recorded in numerical values. There was not enough data to compare the prone protocols with complications related to prone positioning.
The total number of ventilated days was found to be insignificantly more in the supine than in the prone position. Furthermore, the patients' length of stay in the intensive care unit (ICU) was insignificantly longer (16.5 to 26.6 days) in the prone position than in the supine position (19.4 to 24.5 days). The mortality for patients in the clinical trials included was 33.5%.
EVIDENCE-BASED NURSING GUIDELINES
Clinical guidelines have been defined as "systematically developed statements to assist both practitioner and patient decisions in specific circumstances" (Eccles & Mason, 2001:1). The following guidelines have been compiled by the researcher from the results and evidence obtained through summarising all available data and quantitative analysis. Nursing guidelines according to the nursing process are presented in Table 2 as actions, supported by evidence.
Prone positioning shows significant increases in the oxygenation outcomes of patients with ARDS or acute lung injury. Complications with this procedure may be numerous, but these did not occur to the extent that was expected. Nursing interventions that benefited the patient and reduced complications (according to the best available evidence) are described in Table 2.
This study has several limitations. The specific objective of this study was to obtain evidence representative of the nursing activities related to prone positioning. Nursing-care related articles were few and selected articles presented different treatment strategies and outcomes, which mostly did not include nursing activities. A further limitation of this review is the number of studies of poor methodological quality. Most of the studies that were included in this review were of a short duration and thus no long-term benefits or complications of the prone position could be assessed.
Comparisons were made between the supine and the prone positions. Different outcomes were measured, which included oxygenation outcomes, responder and non-responder groups, haemodynamic outcomes, complications in the prone position, mortality, the length of stay in ICU and the total number of ventilated days. These outcomes were measured against the existing prone protocols in order to provide an answer to the research question. The results that were analysed presented the best available evidence to support nursing guidelines for prone positioning of adult, ventilated patients.
The studies that were included in this review reflect mostly medical treatment and the outcomes based on this treatment. The researcher would recommend further research (nursing research) that addresses nursing actions and outcomes regarding patients in the prone position.
ACUTE RESPIRATORY DISTRESS SYNDROME FOUNDATION (n.d.). Available from: http://www.ardsusa.org/ (Accessed 9 October 2006).
ARDS See ACUTE RESPIRATORY DISTRESS SYNDROME. BALAS, MC 2000: Prone positioning of patients with acute respiratory distress syndrome: Applying research to practice. Critical Care Nurse, 20(1):24-37.
BEERS, MH & BERKOW, R eds. 1999: The Merck manual of diagnosis and therapy. Whitehouse Station: Merck Research Laboratories.
BERMAN, NG & PARKER, RA 2002: Meta-analysis: Neither quick nor easy. BMC Medical Research Methodology, 2(1):10. Available from: http://www.biomedcentral.com/1471-2288/2/10 (Accessed 7 March 2005).
BLANCH, L; MANCEBO, J; PEREZ, M; MARTINEZ, M; MAS, A; BETBESE, AJ; JOSEPH, D; BALLUS, J; LUCANGELO, U & BAK, E 1997: Short-term effects of prone position in critically ill patients with acute respiratory distress syndrome. Intensive Care Medicine, 23(10):1033-1039.
BRIGGS, J 2001: Changing practice, evidence based practice information sheets for health professionals; An introduction to systematic reviews, supplement 1. The Joanna Briggs Institute for Evidence Based Nursing and Midwifery. Available from: http:/ /www.joannabriggs.edu.au (Accessed 11 August 2005).
BURNS, N & GROVE, SK 2001: The practice of nursing research: Conduct, critique, & utilization; 4th edition. Philadelphia: WB Saunders.
CHATTE, G; SAB, J; DUBOIS, J; SIRODOT, M; GAUSSORGUES, P & ROBERT, D 1997: Prone position in mechanically ventilated patients with severe acute respiratory failure. American Journal of Respiratory Critical Care Medicine, 155(2):473-478.
CHIUMELLO, D; CRESSONI, M; RACAGNI, M; LANDI, L; BASSI, GL; POLLI, F; CARLESSO, E & GATTINONI, L 2006: Effects of thoracopelvic supports during prone position in patients with acute lung injury/acute respiratory distress syndrome: A physiological study. Critical Care, 10(3):R87.
COOK, DJ; MULROW, CD & HAYNES, B 1997: Systematic reviews: Synthesis of best evidence for clinical decisions. Annals of Internal Medicine, 126(5):376-380.
CURLEY, MAQ 1999: Prone positioning of patients with acute respiratory distress syndrome: A systematic review. American Journal of Critical Care, 8(6):397-406.
ECCLES, M & MASON, J 2001: How to develop cost-conscious guidelines. Health Technology Assessment, 5(16):1-69.
FERGUSON, ND; MEADE, MO; HALLETT, DC & STEWART, TE 2002: High values of the pulmonary artery wedge pressure in patients with acute lung injury and acute respiratory distress syndrome. Intensive Care Medicine, 28(8):1073-1077.
FRIDRICH, P; KRAFFT, P; HOCHLEUTHNER, H & MAURITZ, W 1999: The effects of long-term prone positioning in patients with trauma-induced adult respiratory distress syndrome. Anaesthesia Analgesia, 83(6):1206-1211.
GAINNIER, M; MICHELET, P; THIRION, X; ARNAL, J-M; SAINTY, J-M & PAPAZIAN, L 2003: Prone position and positive end-expiratory pressure in acute respiratory distress syndrome. Critical Care Medicine, 31(12):2719-2726.
GATTINONI, L; TOGNONI, G; PESENTI, A; TACCONE, P; MASCHERONI, D; LABARTA, V; MALACRIDA, R; GIULIO, P; FUMAGALLI, R; PELOSI, P; BRAZZI, L & LATINI, R 2001: Effect of prone positioning on the survival of patients with acute respiratory failure. The New England Journal of Medicine, 235(8):568 574.
GLASZIOU, P; IRWIG, L; BAIN, C & COLDITZ, G 2001: Systematic reviews in health care. A practical guide. Cambridge: University Press.
GOETTLER, CE; PRYOR, JP; HOEY, BA; PHILLIPS, JK; BALAS, MC & SHAPIRO, MB 2002: Prone positioning does not affect cannula function during extracorporeal membrane oxygenation or continuous renal replacement therapy. Critical Care, 6(5):452455.
GROSSMAN, S & BAUTISTA, C 2002: Collaboration yields costeffective, evidence-based nursing protocols. Orthopedic Nursing, 21(3):30-37.
GUERIN, C; GAILLARD, S; LEMASSON, S; AYZAC, L; GIRARD, R; BEURET, P; PALMIER, B; VIET LE, Q; SIRODOT, M; ROSSELLI, S; CADIERGUE, V; SAINTY, J-M; BARBE, P; COMBOURIEU, E; DEBATTY, D; ROUFFINEAU, J; EZINGEARD, E; MILLET, O; GUELON, D; RODRIGUEZ, L; MARTIN, O; RENAULT, A; SIBILLE, J-P & KAIDOMAR, M 2004: Effects of systematic prone positioning in hypoxemic acute respiratory failure: A randomized controlled trial. Journal of the American Medical Association, 292(19):2379-2387.
HAMMERSLEY, M 2002: Systematic or unsystematic, is that the question? Some reflections on the science, art, and politics of reviewing research evidence. Presentation Public Health Evidence Steering Group of the Health Development Agency. Available from: http://www.hda.nhs.uk/evidence/ sys_unsys_phesq_hammersley.pdf (Accessed 6 April 2006).
HEDIN, A & KALLESTAL, C 2004: Knowledge-based public health work. Part 2. Handbook for compilation of reviews on interventions in the field of public health. National Institute of Public Health. Available from: http://www.fhi.se (Accessed 6 April 2005).
HETHERINGTON, J 2004: Newcomers' guide. The Cochrane Collaboration. Available from: file://E:\Cochrane Collaboration newcomers'guide.htm (Accessed 17 June 2005).
JADAD, AR; MOORE, RA; CARROLL, D; JENKINSON, C; REYNOLDS, DJ; GAVAGHAN, DJ & MCQUAY, HJ 1996: Assessing the quality of reports of randomized clinical trials: Is blinding necessary? Controlled Clinical Trials, 17(1):1-12.
JOHANNIGMAN, JA; DAVIS, K; MILLER, SL; CAMPBELL, RS; LUCHETTE, FA; FRAME, SB & BRANSON, RD 2001: Prone positioning and inhaled nitric oxide: Synergistic therapies for acute respiratory distress syndrome. The Journal of Trauma: Injury, Infection, and Critical Care, 50(4):589-596.
JOLLIET, P; BULPA, P & CHEVROLET, J 1998: Effects of the prone position on gas exchange and hemodynamics in severe acute respiratory distress syndrome. Critical Care Medicine, 26(12):1977-1985.
MCAULEY, DF 2001: Guidelines for the use of prone ventilation in acute respiratory distress syndrome and acute lung injury. A dissertation submitted for the intercollegiate Board Diploma in Intensive Care Medicine, Belfast.
MENTZELOPOULOS, SD; ZAKYNTHINOS, SG; ROUSSOS, C; TZOUFI, MJ & MICHALOPOULOS, AS 2003: Prone position improves lung mechanical behaviour and enhances gas exchange efficiency in mechanically ventilated chronic obstructive pulmonary disease patients. Anaesthetic Analgesic, 96(6):1756-1767.
MOHER, D; COOK, DJ; EASTWOOD, SL; OLKIN, I; RENNIE, D & STROUP, DF 2000: Improving the quality of reports of meta-analyses of randomised controlled trials: The QUOROM statement. British Journal of Surgery, 87(11):1448-1454. Available from: http:/ /www.bjs.co.uk (Accessed 5 October 2004).
MULROW, CD 1994: Systematic reviews: Rationale for systematic reviews. Education and debate. British Medical Journal, 309(6954):597-599.
OFFNER, PJ; HAENEL, JB; MOORE, EE; BIFFL, WL; FRANCIOSE, RJ & BURCH, JM 2000: Complications of prone ventilation in patients with multisystem trauma with fulminant acute respiratory distress syndrome. Journal of Trauma: Injury, Infection and Critical Care, 48(2):224-228.
PHILLIPS, B; BALL, C; SACKETT, D; BADENOCH, D; STRAUS, S; HAYNES, B & DAWES, M 2001: Levels of evidence and grades of recommendation. Oxford Centre for Evidence-based Medicine. Available from: http://www.cebm.net (Accessed 7 December 2005).
REVIEW MANAGER (RevMan) 2003: Version 4.2.8 for Windows. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration.
STAMM, K; LAWRENCE, V & RICHARDSON, S 1998: Getting started in systematic reviews. Semi-annual newsletter. VA Center of Excellence. San Antonio and Charleston. Available from: http:// www.sanantonio-asa.uthscsa.edu/pub/downloads/ verdict_98fall.pdf (Accessed 6 April 2006).
THELAN, LA; URDEN, LD; LOUGH, ME & STACY, KM 1998: Critical care nursing; diagnosis and management; 3rd edition. St. Louis: Mosby.
WIKIPEDIA (n.d.) Acute respiratory distress syndrome. Available from: http://www.en.wikipedia.org/wiki/ Acute_respiratory_distress_syndrome (Accessed 10 October 2006).
WIKIPEDIA (n.d.) Pancuronium. Available from: (http:// en.wikipedia.org/wiki/Pancuronium) (Accessed 10 October 2006)
Senior Registered Nurse, Cardio-thoracic Intensive Care Unit
Dr Elsabe Nel
Senior Lecturer, Department of Nursing Science, University of Johannesburg
Prof. Anna Nolte
Professor, Department of Nursing Science, University of Johannesburg
Corresponding author: email@example.com
Table 1: Complications that occurred in the prone position Complications % Decrease in pulse oximetry (Sp[O.sub.2]) 28.0 New pressure sores 25.7 Bradycardia 10.2 Cardiac arrest 10.0 Decreased blood pressure 10.0 Dislodged endotracheal tube 6.3 Facial oedema 5.7 Atelectasis 3.5 Pneumothorax 3.2 Blocked endotracheal tube 3.0 Mild cutaneous and mucosal damage 0.8 Intolerance to enteral feeding 0.8 Loss of venous access 0.4 SVT/tachycardia 0.1 Contractures 0.1 Corneal ulceration 0.1 Loss of arterial access 0.1 Table 2: Evidence-based nursing guidelines according to the nursing process (regarding assessment, planning, implementation and evaluation) ASSESSMENT ACTION EVIDENCE Assess the patient's There is a significant improvement in the indications and Pa[0.sub.2]/Fi[O.sub.2] ratio with patients contra-indications in the prone position. Prone positioning to prone positioning. has been indicated in the acute respiratory distress syndrome (ARDS) with Pa[0.sub.2]/ Fi[O.sub.2] < 200 mmHg. In 45 clinical trials patients with the following conditions were not turned to the prone position: diseases that relate to severe cardiac failure, acute myocardial infarction, cerebral oedema, decreased cerebral perfusion pressure, increased intracranial pressure, haemodynamic instability, spinal instability and fractures, chronic obstructed pulmonary disease, abdominal surgery, open abdomen (peritonitis) and ascitis, cranial trauma and craniofacial surgery, unstable long bone fractures, flail chest or chest wall abnormalities, some infective pulmonary conditions, orthopaedic traction, recent sternotomy, acute bleeding, haemoptysis and alveolar haemorrhage, extensive skin lesions, bone marrow transplants, lung transplants, liver transplants and pregnancy. Assess the patient's Sedation together with paralysis need for sedation or of the patient was more often used paralysis. as part of the prone protocols than sedation only (28 vs 8 trials). Intracranial pressure (ICP) increased with a WMD of 10.2 and central venous pressure (CVP) increased by 0.55 mmHg in patients who were only sedated. Heart rate increased by a WMD of 3.06 beats minute in sedated and paralysed patients. PaC[0.sub.2] decreased with a WMD of -0.61 in patients who were only sedated. Pa[0.sub.2] increased significantly in patients who were both sedated and paralysed. Assess the patient's Only 14 of 1 626 patients showed nutritional needs intolerance to enteral feeding. and determine by which Both parenteral and enteral feeding feeding method the were used in prone protocols. Only patient would benefit. nine studies on prone positioning mention nutrition and feeding. Enteral feeding was given continuously in five studies. Assess each individual The endotracheal tube was blocked in 3% of patient's need for patients. In-line suctioning catheters can airway clearance be placed to ensure easier access to (suctioning). clear the airway of secretions. Assess each patient A total of 25.7% of the patients for skin condition developed new pressure sores and integrity before during prone positioning. and after prone positioning. Assess the stability Patients may become more unstable with of the patient's the turning procedure from supine to condition using prone position. This can be confirmed haemodynamic by the complications that were found parameters and in this review. A decrease in pulse oxygenation outcomes. oximetry (SP02) represented 28% of patients. Tachycardia or supra-ventricular tachycardia occurred in 0.1 of patients. In one study 175 patients had cardiac arrest (10%), bradycardia represented 10.2% of the population and 283 patients out of 791 had decreased blood pressure (10%). PLANNING ACTION EVIDENCE Plan a safe Complications that occur in prone environment (e.g. positioning can have life-threatening emergency trolley consequences that require immediate within reach) to action. The evidence shows that the ensure fast endotracheal tube was dislodged in action in case 6.3% of patients, which would require of life- immediate re-intubation. Loss of threatening venous access occurred in seven events. patients, and loss of arterial access in one patient. These patients are mostly dependent on inotropic support through invasive lines, which, when accidentally removed, can cause a critical event. Plan for five people The evidence shows that the endotracheal (ensure that someone tube was dislodged in 6.3% of patients, who can intubate is which would require immediate re- near or within reach) intubation. The evidence suggests to assist in the that three to six people should be prone-positioning involved in the turning process. process. According to the results, the ideal number of personnel would be 4.6 (rounded off to 5) people. Heart rate increased and the mean arterial pressure (MAP) decreased if only three people assisted in the turning process. Plan in advance for Additional turning and positioning specialised beds, equipment or a pressure-relieving mattresses or turning aid was used in 15 studies. In 9 equipment, e.g. of 15 studies, patients were placed air-cushioned beds on air-cushioned beds, which have or Vollman prone special air mattresses with a positioner. dynamic cell design and automatic adjustment for patient weight (Gainnier, Michelet, Thirion, Arnal, Sainty & Papazian, 2003:2720). Plan the length of The minimum duration in the prone position time for which the was 6.4 hours and the maximum duration was patient will be in 20.7 hours. Oxygenation outcomes were the prone position positive for patients who remained in according to the the prone position for less than 12 hours. physician's These results were not significant. prescription Haemodynamic outcomes varied in their and the patient's results. Heart rate and MAP increased tolerance of the with the patient in the prone position position. for less than 12 hours, while the mPAP (mean pulmonary artery pressure) lowered during this time, which can be considered a positive response. New pressure sores developed more in patients who were in the prone position for more than 12 hours. Mortality was less in patients who remained in the prone position for a shorter period. From this evidence, more favourable results were obtained in patients who remained in the prone position for a period less than 12 hours. IMPLEMENTATION ACTION EVIDENCE Organise the five The person at the patient's head should be staff members as responsible for tracheal intubation. follows: one at Someone else should be responsible for the patient's the patency of the central lines (Chatte, head and two on Sab, Dubois, Sirodot, Gaussorgues & either side of Robert, 1997:474). the patient's body. Choose an appropriate Volume-controlled mode of ventilation mode of ventilation was used in 14 prone protocols and the and individualise pressure-controlled mode of ventilation ventilator settings. was used in 9 trials. The Pa[0.sub.2], Pa[O.sub.21]Fi[O.sub.2] ratio and minute volume showed better results in the volume-controlled mode of ventilation. These results were not conclusive, however, owing to the lack of data on the pressure- controlled mode of ventilation. Throughout study protocols, the mean Fi[0.sub.2] (fraction of inspired oxygen) was 0.8%, the mean frequency setting was 18.4, tidal volume (measured as mi/kg) was 8.6 ml/kg, and the maximum PEEP (Positive end expiratory pressure) applied was 11.1 mmHg. Monitor the patient's 28% of patients presented with a decrease oxygenation outcomes in SP[0.sub.2] and 25% of patients can be by performing blood classified as non-responders. 75% of gas sampling before patients responded positively to prone and after prone positioning with significant increases positioning and in Pa[0.sub.2]. Blood gas sampling was regularly on an regularly performed in all studies to individual basis. measure the outcome and effect of treatment. The effects and outcomes of prone positioning should also be assessed when this treatment is used. Aspirate gastric Only two studies mentioned that content before gastric content was aspirated prone positioning (and in both cases it was done and continue to measure residual volumes). with appropriate nutrition after the procedure. Secure the patient's Only a few studies mentioned suctioning endotracheal tube practices (clearing of the airway), with before prone no mention of the frequency and actual positioning. suctioning methods. Suctioning was mostly In-line suctioning performed before and after positioning catheters can be (Blanch, Mancebo, Perez, Martinez, Mas, placed to ensure Betbese, Joseph, Ballos, Lucangelo & easier access Bak 1997:1034; Chatte et al. 1997:474; to clear the Johannigman, Davis, Miller, Campbell, airway of Luchette, Frame & Branson, 2001:590). secretions and Closed suctioning systems were used suction when (Fridrich, Krafft, Hochleuthner & necessary. Mauritz, 1999:1207). Disconnect monitoring Vital signs were monitored closely devices or lines before and after turning events in that could obstruct most studies, for example in the or restrict the study by Goettler, Pryor, Hoey, turning of a Phillips, Balas and Shapiro patient. Ensure (2002:453). Evidence from the the patient's trials included suggests that safety at all the patient should be removed times or rearrange from certain monitoring devices monitoring cables, to facilitate an unrestricted probes and turn (Mentzelopoulos, Zakynthinos, invasive monitoring Roussos, Tzoufi & Michalopoulos, devices before the 2003:1760). procedure to ensure continuous monitoring. Monitor haemodynamic All the haemodynamic parameters increased parameters before, when the patient was turned from the during and after the supine to the prone position. Pulmonary prone positioning artery wedge pressure increased process for early significantly. In order to monitor detection of any patients effectively, these patients had instability. indwelling catheters (pulmonary artery, arterial and venous), pulse oximetry probes, electrocardiographic leads, Foley catheters and a few other devices (e.g. cardiac output monitors). In all 45 trials included, these devices were used to some extent to facilitate continuous measurements and monitoring. Turning the patient into the prone position and care provided in this position Option 1: Position the Eleven studies supported a lateral patient's head position (the patient's head was laterally, or turned laterally, supported under Option 2: Position the the side of the head). Theoretically patient's head the lateral positioning of a facing downward (on patient's head may compress the forehead and chin). jugular veins, which can cause raised intracranial pressure. Seven Provide pressure studies supported a full frontal relief under pressure position (head in total prone parts, e.g. the position and supported under the patient's ears, forehead and chin). In the study forehead or chin, by Jolliet, Bulpa and Chevrolet and to avoid eye (1998:1980), three patients developed injuries. pressure sores on the chin. The evidence found one patient with an infectious corneal ulceration. There was no evidence suggesting that one position was preferred over the other. Ensure the patency and The loss of venous access occurred in correct position of seven patients and the loss of arterial invasive lines and access in one patient. tubes. Apply a moisture barrier Most patients presented with mild to the patient's cutaneous damage or facial oedema to entire face to some degree. Unfortunately these protect the skin complications were not measured or integrity against recorded in numerical values. Care secretions from the should be taken to avoid these nose and mouth. complications. Option 3: Turn the Nine studies supported option 3 and patient without 13 studies option 4. Oxygenation any support under improved markedly with option 3. the chest or In the group without support under pelvis (the the chest and pelvis (option 3), patient lies flat the Pa[0.sub.2]/Fi[O.sub.2] ratio against the showed a significant increase with mattress), or a WMD of 61.02 (95% CI 32.67; 89.38) Option 4: between the supine and prone Support the positions. patient's body under the chest and pelvis (with pillows or other pressure- relieving materials) with the patient's abdomen protruding and hanging free from the bed. Option 5: Keep the In 12 studies the patients' arms patient's arms were placed next to their bodies straight and (parallel) and in four studies the parallel to the patients' arms were bent at the body, or shoulder and at 90[degrees] at the Option 6: Bend one elbow. arm at the shoulder and 90[degrees] at the elbow. Provide movement to Head and arm positions were changed extremities on a two hourly in studies that reported regular basis and mobilisation of extremities, for turn the patient's example in the study by Offner, head and arms two Haenel, Moore, Biffl, Franciose hourly to the and Burch (2000:225). Three patients opposite position. presented with contractures. Position the patient's According to the evidence, body into left and 4.4 hours were used in the right lateral supine position for procedures decubitus position and nursing care. In the study by for nursing care, Guerin, Gaillard, Lemasson, or turn the patient Ayzac, Girard, Beuret, Palmier, back to the Viet Le, Sirodot, Rosselli, supine position. Cadiergue, Sainty, Barbe, Combourieu, Debatty, Rouffineau, Ezingeard, Millet, Guelon, Rodriguez, Martin, Renault, Sibille and Kaidomar (2004:2381), it was suggested that patients be turned to a left and right lateral decubitus position. Record the patient's In 45 studies 25% were non-responders and response or non- 75% were responders to prone positioning. response to this Patients who first responded generally procedure. It is maintained or increased their oxygenation suggested that response. Patients who did not respond patients be monitored initially may respond at a second or third meticulously attempt. A patient was considered a for the first responder when the Pa02/Fi[O.sub.2] ratio 30 minutes to an increased by at least 15 to 20% or more, hour to assess or when there was an improvement in their response Pa[0.sub.2] of more than 7 to 10 mmHg or to the procedure. 20% after 30 minutes. There were a few variations on these classification criteria which included an increase of 20 mmHg in the Pa[0.sub.2] /Fi[O.sub.2] ratio or a decrease of more than 10% in the oxygenation index. Monitor the patient's In the study by Chiumello, Cressoni, intake and urine Racagni, Landi, Bassi, Polli, Carlesso output. and Gattinoni (2006:6), urine output was less in the prone (mean 90; [+ or -]SD 64) than in the supine (mean 111; [+ or -] SD 71) position. EVALUATION ACTION EVIDENCE Evaluate the patient's The overall shunt fraction in patients oxygenation response decreased from 34.3% in the supine to to the prone position 26.6% in the prone position. Pa[0.sub.2] and follow up on non- increased from 89.3 mmHg in the supine responders. to 137.2 mmHg in the prone position. Evaluate the Mild cutaneous and mucosal damage occurrence of occurred in 14 patients complications and I and facial oedema in 93 cases. take appropriate nursing actions to prevent these complications. Evaluate the patient's The mean number of prone cycles was condition and 10.6 times, ranging from 1-23 times. determine whether Oxygenation outcomes improved in one turn is patients who were turned only once. sufficient or The dislodgement of endotracheal whether more turns tubes and the number of new pressure should be performed. sores occurred more often in patients with two or more turning events.