Identifying the site for intercostal catheter insertion in the emergency department: Is clinical examination reliable?

Transcription

1 bs_bs_banner Emergency Medicine Australasia (2014) 26, doi: / ORIGINAL RESEARCH Identifying the site for intercostal catheter insertion in the emergency department: Is clinical examination reliable? Peter CARTER, 1 Sheree CONROY, 1,2 Jade BLAKENEY 1 and Bimal SOOD 1,2 1 Emergency Department, Toowoomba Hospital, Toowoomba, Queensland, Australia, and 2 Rural Clinical School, University of Queensland, Toowoomba, Queensland, Australia Abstract Objective: To determine whether ED doctors, comprising both consultants and registrars, can accurately identify the 4th or 5th intercostal space (ICS), commonly used for intercostal catheter insertion. Methods: An observational study was designed using a sample of ED doctors applying their clinical skills to a convenience sample of patients reflecting a heterogeneous mix of ED patients. Patients already receiving a CXR in our ED were examined by a registrar or consultant who placed a radiopaque marker on the patients chest wall over the site they determined to be the 4th or 5th ICS. Consultant radiologists reported the marker s position from postero-anterior projection CXRs, and results were analysed comparing consultants with registrars, right to left hemithoraces and male to female patients. Results: ED doctors participating in the present study placed the marker over the 4th or 5th ICS 36.2% of the time, with no significant difference between consultant and registrar groups, nor right or left hemithoraces. Accuracy was improved in female patients compared with male patients. Conclusion: Emergency registrars and consultants sampled from a regional ED appeared unable to reliably identify the 4th or 5th ICS, as evidenced by marker position, in a heterogeneous patient population. Key words: accuracy, chest drain, clinical examination, intercostal catheter, intercostal space. Introduction Intercostal catheter (ICC) insertion is a common procedure in emergency medicine. They are inserted to treat many diseases, including traumatic haemothorax and pneumothorax, spontaneous and iatrogenic pneumothorax, and some pleural effusions. Physicians rely on their clinical skills to identify the relevant anatomical landmarks, ensuring safe insertion. Serious complications might arise from insertion of an ICC through an incorrect intercostal space (ICS). The most commonly targeted ICC insertion point is Bulau s position, the 4th or 5th ICS in the mid to anterior axillary line. 1 This lies within the triangle of safety bordered by the anterior edge of latissimus dorsi, the lateral edge of pectoralis major and a line superior to Correspondence: Dr Peter Carter, c/o Intensive Care Unit, The Townsville Hospital, Douglas, QLD 4814, Australia. peterj.carter@health.qld.gov.au Peter Carter, MBBS, Grad Cert Clinical Education, Emergency Registrar; Sheree Conroy, MBBS, FACEM, DEMT, Grad Cert Clinical Education, Emergency Physician, Senior Lecturer, Director of Clinical Training; Jade Blakeney, MBBS, BBiotech (Hons), PhD, Emergency PHO; Bimal Sood, MD, FRANZCR, Senior Lecturer, Director of Radiology. Accepted 22 July 2014 Key findings Emergency doctors were unable to reliably identify the 4th or 5th intercostal space using clinical examination skills. Seniority did not appear to improve reliability. Reliability was improved in male compared to female patients. the horizontal level of the nipple with the apex below the axilla. 2 Having seen malpositioned ICCs lead to adverse events, the authors designed a study to determine whether ED doctors can accurately identify the 4th or 5th ICS using their clinical skills alone in a heterogeneous group of patients. A review of the literature was performed, but did not sufficiently answer this question. A 2005 study 3 assessed the knowledge and accuracy of 25 ED doctors in naming and locating the correct site for needle thoracentesis, deemed to be the 2nd ICS in the mid-clavicular line, with a single attempt. Of this group comprising senior house officers, registrars and consultants, only one correctly named and identified the 2nd ICS in the mid-clavicular line. Of the remainder, 14 located a site too medial, whereas 10 were outside of the 2nd ICS altogether. A 2010 study 2 asked 50 junior doctors in the UK to locate the triangle of safety on a photograph of the lateral chest wall. They found that only 44% of participants correctly identified and located the triangle of safety, with the remainder indicating a site

2 ARE CLINICAL EXAMINATION SKILLS RELIABLE? 451 below the nipple line. At full expiration, the diaphragm contacts the costal portion of the parietal pleura inferiorly from the 7th rib, 4 worryingly close to the inferior margin of the triangle of safety. A prospective audit of ICCs inserted at a level 1 trauma centre in Australia found that 1% was inadvertently inserted into the peritoneum. 5 Injury to the liver, spleen and stomach has resulted from intra-abdominal ICC insertion, 6,7 with a survey of major complications associated with ICC insertion in the UK in 2009 finding that, of 37 misplaced drains, 10 were placed in the liver, six in the intraperitoneal space without organ injury, five in the spleen and one in the colon. 8 Unfortunately, there is a paucity of literature examining factors contributing to incorrect ICC placement. Following review of the literature, the authors hypothesised that ED doctors would not be accurate in identifying the 4th or 5th ICS. Although the chance of incorrect insertion of an ICC appears relatively low, it is hoped that determining the reliability of doctors clinical skills in locating the site used for ICC placement will improve patient safety, and stimulate discussion around future education and training initiatives. Methods Aims Primary aim: To determine whether ED doctors can reliably identify the 4th or 5th ICS on the external thoracic wall. Secondary aims: To explore the effects that seniority of doctor and patient sex might have on ED doctors accuracy. The position of markers placed on the right and left hemithoraces was compared to assess the internal validity of our results. Study design An observational study was designed using ED doctors applying their clinical skills to a convenience sample of patients reflecting a heterogeneous mix of ED patients. The present study was carried out in the ED of Toowoomba Hospital, a TABLE 1. Patient inclusion and exclusion criteria Inclusion criteria Already undergoing CXR for investigation and management in the ED Written consent obtained regional referral centre in South-East Queensland, 110 km west of Brisbane. The ED triages presentations annually and, during 2013, was staffed by 11 FACEMs sharing eight full-time equivalent (FTE) consultant positions and 14 registrars sharing 11.5 FTE positions. Participants were enrolled from August to December Ethical approval was granted by the Darling Downs Hospital and Health Services Human Research Ethics Committee HREC/13/QTDD/34. Emergency consultants and registrars working during the enrolment period were invited to participate. Other doctors were able to enrol patients, but did not perform the examination to identify the intercostal space. Before the study commencing, all registrars and consultants working in the ED were offered training regarding the projects aims, purpose and their role in the research. Participating doctors were reassured that all results would be anonymous and their individual performance would not be assessed. Patients who required a CXR for their evaluation in the ED were invited to participate in the study. Patient inclusion and exclusion criteria are displayed in Table 1. Once written consent was obtained, the patient was taken to the ED radiology suite and positioned for their CXR, with their arm abducted and internally rotated, to simulate the position commonly adopted for ICC insertion. An ED registrar or consultant was asked to examine the patient while in the ED radiology suite to identify the 4th or 5th ICS. The method of examination was not stipulated, with participants instructed to follow their usual practice. Examining clinicians placed a radiopaque marker on the patient s skin overlying their selected site. CXRs were taken immediately following marker placement, with patients instructed to hold their arm in the examined position. Subjects whose arm position on CXR suggested that they had likely lowered their arms after marker placement were further excluded from the results. Each CXR was reviewed by a consultant radiologist to determine which ICS the marker was placed over. Two experienced consultant radiologists working at Toowoomba Hospital reported all results from erect posteroanterior projection CXR to improve the precision of identifying the corresponding ICS radiologically. Reporting radiologists were blinded to the identity and seniority of the examining clinician. Convenience sampling of both doctors and patients occurred as determined by doctors availability and the willingness of patients to be involved. Statistics Exclusion criteria Unwilling to consent Unable to consent because of lack of capacity (i.e. altered level of consciousness, intoxicated etc.) Age < 18 years Pregnant No ED doctor available to examine and place marker immediately before CXR because of clinical duties Trauma, or other medical condition, rendering the patient unable to be positioned for postero-anterior projection CXR without the need for sedation An initial sample size of 385 patients was calculated using a 0.5 margin of

3 452 P CARTER ET AL. error, 95% confidence interval and accepting a standard deviation of 0.5. The sample size was reviewed after approximately 20% of the estimated sample had been enrolled and the sample size was readjusted, with enrolment ceasing after 116 included patients. Statistics were calculated with the use of Prism, version 6, by GraphPad Software Inc. (San Diego, CA, USA). Contingency tables were used to review correct and incorrect identification of the 4 5th ICS by experience of doctor (registrar vs consultant) and sex of patient. Contingency tables were used to analyse correct identification of 4th to 5th ICS when data were grouped by experience (registrar vs consultant). Confidence intervals were calculated using the modified Wald method, with Fisher s analysis then applied to TABLE 2. Age (years) Demographics of enrolled patients Total number determine the significance of any differences between study groups. Results All 14 registrars participated in the project. These doctors had varied experience levels ranging from postgraduate year 3 prevocational doctors to advanced trainees of the ACEM. The group comprised roughly 30% advanced ACEM trainees, 55% provisional ACEM trainees, with the remainder rural generalist or general practice trainees seeking extra skills in emergency medicine. Of the consultant group, nine of the 11 FACEMs participated in the study. The demographics of enrolled patients are summarised in Table 2. One hundred and forty-four patients were approached for enrolment. The flow of participants through Number of men Number of women Figure 1. Excluded n = 28 Total number of patients enrolled n = 144 Correct position n = 42 Included n = 116 Flow of participants throughout the present study. Incorrect position n = 74 the study is shown in Figure 1. Of the 28 exclusions, one patient had not met inclusion criteria after not eventually receiving a CXR; one patient was pregnant; 23 did not have markers placed on their chest wall because of workforce limitations; and three patients were further excluded from analysis as their arms were poorly abducted on the CXR, suggesting that they might have been lowered after the dot was placed. The group of ED doctors studied placed the marker over the 4th or 5th ICS 36.2% of the time. Tabulated results are shown in Table 3. Diagrammatical breakdown of marker positions is depicted in Figure 2. Markers were most often incorrectly placed over the 6th and 7th ICS (27.5% and 20.7%, respectively). Although there is a large difference in the number of consultantplaced markers compared with those placed by registrars, there was no significant difference in proportion of correct marker position between the groups (P value = ). There was also no significant difference in the number of correct marker positions on the right compared with the left hemithorax (P value = ). The number of correctly positioned markers was significantly improved on female patients (50% in correct position) when compared with male patients (27%), with a P value of Discussion The results of the present study showed that registrars and consultants working in our regional ED were unable to reliably place a marker over the 4th or 5th ICS. Inability of ED doctors to correctly identify the 4th or 5th ICS might lead to increased patient morbidity, particularly if placed inferior to the diaphragm. 5 7 The present study has demonstrated a preponderance to identify a site for potential ICC insertion inferior to Bulau s position, approaching the point at which the diaphragm is in direct contact with the costal portion of the parietal pleura in expiration. 4 This might lead to increased risk of intraperitoneal placement, particularly when insertion is performed in the supine position where increased intra-abdominal pressure might further elevate the diaphragm

4 ARE CLINICAL EXAMINATION SKILLS RELIABLE? 453 TABLE 3. Tabulated results Total placed (i.e. a multiply injured trauma patient). These results support other authors findings regarding the poor accuracy of clinical skills employed to identify thoracic anatomical sites used for invasive emergency medicine procedures. 2,3 The ED in which the study was performed employs registrar level staff from a variety of vocational training backgrounds. Limiting participants to ACEM advanced trainees and FACEMs might have improved the generalisability of the study, but would not reflect the real world group of doctors in our ED who are expected to insert ICCs independently. It is plausible that the accuracy of advanced Inside target zone (%) 95% confidence interval for % of markers inside target zone Outside target zone (%) Total (36) 74 (64) Seniority of doctor Consultant 16 5 (31) [11 51] 11 (69) Registrar (37) [28 46] 63 (63) Side of patient Right side (34) [23 45] 44 (66) Left side (39) [26 52] 30 (61) Sex of patient Male (27) [17 37] 51 (73) Female (50) [37 63] 23 (50) 2nd ICS n = 4 3rd ICS n = 4 4th ICS n = 11 5th ICS n = 31 6th ICS n = 32 7th ICS n = 24 8th ICS n = 10 Figure 2. Diagrammatic representation of intercostal space (ICS) marker positions. (Note that although the left hemithorax is depicted, results are for both right and left side of patient.) trainees alone would have been no better than that of the studied registrar group. There was, after all, no significant difference found in the incidence of correct ICS space identification between consultant and registrar level doctors. Admittedly, however, consultants placed only a small number of markers in the present study. The surface anatomy of the right and left hemithoraces do not differ in any way that should alter the ability of a clinician to locate the intercostal spaces. 4 Accordingly, the present study found no significant difference in the number of correctly placed markers on the left compared with the right hemithorax. The accuracy of ED doctors in placing markers over the 4th or 5th ICS in the present study was significantly improved in female patients compared with male patients. The reason for this is unclear. Chest wall thickness was not measured in any patients during the present study, and it would be interesting to explore whether accuracy is altered in patients with differing chest wall thickness. The present study was performed in a single site and the results might not be generalisable to other institutions or specialties. Despite this, it might still be prudent to develop novel techniques, such as the use of US, to ensure safe ICC placement. In 2013, Bhatia and colleagues found that accurate location of the ICS for intercostal nerve block injection was significantly improved by the use of US compared with using anatomical landmarks alone. 9 It is possible that US could similarly be employed in preparing for ICC insertion by ensuring that the selected site is clear of vulnerable structures 7 ; however, there is currently no evidence that the use of US improves the ability to identify a specific ICS in the ED. Limitations The results might have been biased if individual participants monopolised the placement of markers on patients chest walls. Participation in the present study was voluntary and anonymity was maintained in an effort to improve participation, so it is unclear how many dots were placed by individual participants. Doctors were not randomised to examine patients and place markers before their CXR. This was done in an attempt to minimise disruption to clinical care and patient flow. Radiographers were instructed to approach whichever ED registrar or consultant was available to place the marker. Anecdotally, however, all registrar participants each reported placing multiple markers and consultant participants reported placing no more than two markers each. The method of locating the 4th or 5th ICS was not stipulated, so as to determine the reliability of doctors clinical skills used in their everyday practice. All doctors participating in the present study were afforded the opportunity to be reminded of the methods of locating the 4th or 5th ICS to standardise the knowledge base of participants. Conclusion Emergency registrars and consultants sampled from a regional ED appeared unable to reliably identify the 4th or 5th ICS, as evidenced by marker position, in a heterogeneous patient population. Most markers were placed over a site worryingly close to the diaphragm. Further research is warranted to determine the accuracy of the wider ED community in identifying the 4th or 5th ICS. It would also be useful to explore whether poor accuracy translates to an increased risk of ICC

5 454 P CARTER ET AL. misplacement. Education and training might show improvement in the accuracy of doctors clinical skills, although this needs further investigation. Novel techniques for identifying ICC insertion sites might need to be developed to reduce the risk of iatrogenic injury to ED patients requiring ICC insertion. Acknowledgements We thank the staff of Toowoomba Hospital s Emergency and Radiology Departments for their assistance in performing this research. In particular, we thank Dr Joseph for his assistance with reporting the marker positions. Author contributions PC, SC, JB and BS were involved in study design; PC, SC and BS performed data collection; PC and JB were involved in data analysis; PC, SC and JB were involved in manuscript preparation and editing; PC, SC, JB and BS gave final manuscript approval. Competing interests None declared. References 1. Maybauer M, Geisser W, Wolf H, Maybauer D. Incidence and outcome of tube thoracostomy positioning in trauma patients. Prehosp. Emerg. Care 2012; 16: Elsayed H, Roberts R, Emadi M, Whittle I, Shackloth M. Chest drain insertion is not a harmless procedure are we doing it safely? Interact. Cardiovasc. Thorac. Surg. 2010; 11: Ferrie E, Collum N, McGovern S. The right place in the right space? Awareness of site for needle thoracentesis. Emerg. Med. J. 2005; 22: Moore KL, Dalley AF. Clinically Oriented Anatomy, 4th edn. Philadelphia, PA: Lippincott, Williams & Wilkins, Alrahbi R, Easton R, Bendinelli C, Enninghorst N, Sisak K, Balogh Z. Intercostal Catheter Insertion: are we really doing well? ANZ J. Surg. 2012; 82: Kesieme EB, Dongo A, Ezemba N, Irekpita E, Jebbin N, Kesieme C. Tube thoracostomy: complications and its management. Pulm. Med. 2012; 2012; Article ID Wrightson J, Fysh E, Maskell N, Lee Y. Risk reduction in pleural procedures: sonography simulation and supervision. Curr. Opin. Pulm. Med. 2010; 16: Harris A, O Driscoll B, Turkington P. Survey of major complications of intercostal chest drain insertion in the UK. Postgrad. Med. J. 2010; 86: Bhatia A, Gofeld M, Ganapathy S, Hanlon J, Johnson M. Comparison of anatomic landmarks and ultrasound guidance for intercostal nerve injections in cadavers. Reg. Anaesth. Pain Med. 2013; 38: