Nasogastric tubes are widely used in the world’s hospitals, yet in spite of fierce campaigning to expose the dangers, patients are still dying from the complications of wrongful insertion. Sara Williams and MPS medicolegal adviser Dr Gordon McDavid explore how to avoid these risks.
In 2010 75-year-old Maurice Murphy died in hospital as a result of a misplaced nasogastric tube. He was being treated for liver failure and required a nasogastric (NG) tube to be inserted. Unfortunately this ended up in his right lung instead of his stomach and feeding commenced, resulting in fatal pneumonia.
At the inquest it emerged that a junior doctor was challenged by a nurse to confirm that the tube was in the right place. The doctor in question overruled her, saying: “You don’t have a brain to remember that I told you to start the feed as the tube is in the right position.” It also emerged that there was an x-ray flagging the error.1 So why hadn’t anyone seen it?
It would appear that a combination of factors led to the death of Mr Murphy – the misplaced confidence of the junior doctor, the fact the standardised procedure for inserting a tube was not followed, and that the x-ray was not reviewed. NG tubes are commonly used across the world to treat stroke patients with dysphagia or those on ventilators, and are generally accepted as being safe pieces of equipment. They are used in the short to medium term (six weeks); longer term feeding usually requires insertion of gastrostomy or jejunostomy tubes (PEG or PEJ).
A junior doctor was challenged by a nurse to confirm that the tube was in the right place. The doctor in question overruled her
Although feeding by NG tubes is not routinely captured in activity data, in the UK alone around 170,000 tubes are supplied to the NHS each year.
Many practitioners may not have considered the real potential for harm that these innocent-looking plastic tubes may present, particularly if they are misplaced in the patient’s oesophagus or, worse, a bronchus. If not detected before feeding, patients can suffer complications like pneumonia, which can be fatal.
Junior cardiology trainee doctor Dr Owen Bebb inadvertently caused a pneumothorax using an NG tube while working in a busy teaching hospital. “I was bleeped just before my shift ended and asked to check the position of an NG tube my consultant had inserted into a female patient who was nil by mouth due to an unsafe swallow post-stroke. The CXR showed that the tube was in the left bronchus.
“Unfortunately, I had to free the tube from the bridle it had been attached to before removing it. To reinsert the tube I couldn’t use the standard technique of having the patient swallow and so went blindly. The first time it coiled in her mouth, the second time it inserted smoothly without any resistance. As we were unable to aspirate any contents she went for a further chest x-ray to confirm the position.
“I came in after the weekend to find that I had unwittingly caused a pneumothorax and still have no idea how. Fortunately the patient received no lasting damage.”
Risks of nasogastric tubes
The ‘whoosh’ or ‘blow’ test
The UK’s National Patient Safety Agency (NPSA) issued guidance in 2005 highlighting the unreliability of certain tests to detect the placement of NG tubes, such as the ‘whoosh’ test (listening for bubbling sounds after blowing air through the NG tube with a syringe) and pH testing by non-quantitative, coloured litmus paper.2 The NPSA recommend pH testing using pH indicator paper as a first-line check – pH levels between 1 and 5.5 are safe.
Misinterpretation of x-rays
Between 2005 and March 2011 the NPSA was notified of 21 deaths and 79 cases of harm due to misplaced NG tubes. The single greatest cause of harm was due to misinterpretation of x-rays, accounting for about half of all incidents and deaths. A chest x-ray is required if the first-line check fails to prove the NG tube is safe for use.
Flushing nasogastric tubes
The NPSA recently highlighted the deaths of two patients, where staff had flushed NG tubes with water before the initial placement. The mix of water and lubricant gave a pH reading below 5.5, so practitioners assumed that the NG tubes were correctly placed, when they were not.3
According to Sir Liam Donaldson, Patient Safety Envoy for the World Health Organisation, recent findings indicate that NPSA guidance is not being heeded, such as feeding despite obtaining nasogastric aspirates with pH between 6 and 8, instilling water down the tube before obtaining an aspirate, not checking tube placement or not recording written confirmation of such checks.
Sir Liam said: “An NPSA audit suggested great variation among 166 junior doctors at five pilot hospital sites in England and Wales, with low awareness of harm and continued use of checks, such as the ‘whoosh test’ or blue litmus paper, as bad practice. Fewer than a quarter were aware of existing guidance and less than a third of the junior staff had received formal training on x-ray interpretation.
Fewer than a quarter were aware of existing guidance and less than a third of the junior staff had received formal training on x-ray interpretation
“Because of the preventable nature of this harm, last year misplaced nasogastric tubes were confirmed by the Department of Health in England as a ‘never event’, one of a restricted list of serious avoidable events that could incur financial penalties for providers.”4
England and Wales are not alone; other countries such as Malaysia routinely use the ‘whoosh test’ to detect the placement of NG tubes.
Radiograph 1 shows the tip of an NG tube above the diaphragm and on the right-hand side of the thorax. The presence of ECG leads makes the interpretation of the radiograph more difficult.
Radiograph 2 shows the tip of the nasogastric tube apparently below the left hemidiaphragm, but the tube clearly follows the contours of the left bronchus. In fact, the tube is positioned in the left lower lobe of the lung.
Suggested questions to ask yourself when interpreting an x-ray.
- Can you see the tube?
- Does the tube path follow the oesophagus?
- Can you see the tube bisect the carina?
- Can you see the tube cross the diaphragm in the midline?
- Does the tube then deviate immediately to the left?
- Can you see the tip of the tube clearly below the left hemi-diaphragm?
- Does the length of tube inserted suggest it should be in the stomach?
- Does the x-ray cover enough of the area below the diaphragm to see the tube clearly or does it need to be repeated? Note that if the image is not clear, you can manipulate the PACS windows to improve contrast and visualisation.
Case report: Think before you sleep
Mr S was a 70-year-old librarian who had a long history of recurrent colitis due to Crohn’s disease. Despite maximal medical treatment, he experienced recurring symptoms of severe abdominal pain and rectal bleeding, so was admitted to hospital.
Following a period of parenteral steroid therapy, Mr S’s bleeding continued and he required an exploratory laparotomy. Prior to surgery a barium enema revealed a discrete area of abnormal bowel, which was felt to be responsible for his symptoms. It was hoped that the inflamed section of bowel could be surgically resected to alleviate the problems.
Mr S underwent a pre-op assessment by senior anaesthetic trainee, Dr P. He was noted to have a history of angina and COPD, but these chronic conditions were stable.
Dr P had performed this procedure many times before and felt confident to do it independently
On the day of Mr S’s surgery, the operation took place without complication and Dr P inserted an NG tube. As Mr S was intubated, Dr P used a laryngoscope and Magill’s forceps to insert the NG tube. Dr P had performed this procedure many times before and felt confident to do it independently.
During the insertion, Dr P found it difficult to visualise the proximal end of the oesophagus, but based on the smooth insertion assumed the NG tube was in place. Satisfied with the procedure, he escorted Mr S to ICU to begin his recovery.
On arrival in ICU, Dr P still needed to confirm the position of the NG tube. Unable to aspirate fluid, he wanted to auscultate the stomach while instilling air through the NG tube (the ‘whoosh’ or ‘blow’ test) – this was in line with the local protocols at the time. As Mr S had had a laparotomy, Dr P was unable to access the epigastrium to carry out this manoeuvre due to a large wound pad covering the area.
Due to a backlog in the radiology department, the x-ray was not carried out before the end of Dr P’s shift
Dr P decided to arrange a chest x-ray to confirm the position of the NG tube. Due to a backlog in the radiology department, the x-ray was not carried out before the end of Dr P’s shift. Dr P handed over the task of reviewing the film to the nightshift trainee, Dr A. Unfortunately, Dr P failed to inform Dr A that the x-ray was to check the position of the NG tube. Dr P had been so rushed that he had not documented the insertion of the NG tube, or that the correct position had yet to be confirmed.
Following the handover, Dr A noticed there was a leak from Mr S’s endotracheal tube and she injected approximately 1ml of air into the tube’s cuff, which resolved the leak.
Dr A was called away to an emergency, but instructed one of the nurses to observe Mr S. The results of Mr S’s chest x-ray arrived, but Dr A was too busy to review it immediately. When Dr A did look at the x-ray, she glanced at it quickly, verbally informing the nurses that it “looked ok”, referring to the ET position as “satisfactory” and the lungs looking “grossly normal”. She did not document this in the medical notes.
Dr P had been so rushed that he had not documented the insertion of the NG tube, or that the correct position had yet to be confirmed
Unfortunately, Mr S had to return to theatre for an anastomotic leak repair and subsequently required prolonged intubation, blood transfusions, IV fluids and inotropic support after the second surgery. With treatment Mr S’s haemodynamic parameters stabilised although he began to develop renal failure.
On-call consultant anaesthetist Dr W took the decision to begin feeding. During this time, the original NG tube remained in-situ. No-one realised the initial chest x-ray had not been formally reviewed.
When Dr A did look at the x-ray, she glanced at it quickly, verbally informing the nurses that it “looked ok”, referring to the ET position as “satisfactory” and the lungs looking “grossly normal”. She did not document this in the medical notes
About 12 hours later, Mr S’s nurse aspirated feed-like material from his ET tube and feeding was immediately stopped. Dr A was asked to review the patient immediately and while the review was being completed, radiology phoned to advise that the chest x-ray taken before the weekend showed the NG tube was positioned incorrectly.
Despite aggressive treatment for aspiration pneumonitis, unfortunately Mr S’s condition worsened and he died two days later.
The postmortem outlined the cause of death as aspiration pneumonia due to a misplaced nasogastric tube in right main bronchus, left hemicolectomy for intestinal haemorrhage, ischemic heart disease and chronic obstructive airways disease.
Initially, Dr P and the nurses involved were arrested on suspicion of manslaughter and interviewed by the police under caution. However, following an investigation including the obtaining of expert opinion that agreed that the level of care, although suboptimal, did not meet the necessary criteria for a criminal offence, lead to the police deciding that there was insufficient evidence to proceed with the prosecution – the charges were dropped.
The Crown Prosecution Service’s expert opined that the responsibility fell to the practitioner placing the NG tube to check its position. If this is not possible, that practitioner should leave clear written instructions on who will check it.
Two years later, the practitioners involved were called to an inquest and MPS arranged legal representation for Dr P. At the inquest, Dr P accepted that it was an omission on his part not to have specifically recorded in the notes the NG tube insertion.
The coroner recorded a narrative verdict, concluding that Mr S died of aspiration pneumonia due to the incorrect positioning of an NG tube. She criticised all the practitioners involved for failing to detect this.
[The coroner] was satisfied that systems had been implemented by the trust to prevent such an incident recurring
Given that the expert felt that the standard of care was acceptable, however, she took no further action, as she was satisfied that systems had been implemented by the trust to prevent such an incident recurring.
Mr S’s wife launched a claim against the hospital. This was settled for a moderate sum.
Avoiding the risks
Individual clinicians should consider the following:
- Is nasogastric feeding right for this patient? – Seek specialist advice if the patient has a high risk of aspiration or any deviation to normal anatomy, such as pharyngeal pouch, strictures or facial trauma, in which cases fluoroscopic guidance can often be used. The decision to feed should be agreed by two competent professionals and recorded.
- Does this need to be done now? – Risks are greater during the night.
- Am I competent to do this? – Ensure you have had training in safe insertion and checking, including interpretation of x-rays.
- How can I check the right amount of tube has been inserted? – Use “NEX” measurement (by placing exit port of tube at tip of Nose, stretching to Earlobe and then down to Xiphisternum) to guide insertion. The tube length should be confirmed and recorded before each feed to check it has not moved.
- Do I know how to test for correct placement? – Do not flush tubes or start feeding until you can confi rm by testing with quantitative pH indicator paper.
- What is a safe pH level? – Obtain a nasogastric aspirate (pH levels between 1 and 5.5 are safe). Double-check with another person if you are unsure. Always record the result and the decision to start feeding.
- When should I get an x-ray? – If no aspirate can be obtained or the pH reading is above 5.5, request an x-ray specifying the purpose so the radiographer knows the tip of the NG tube should be visible.
- What should I look for on the x-ray? – That the tube is in the correct position (see guide in Figure 1).
- What about repeat checks? – Tubes can be dislodged so they should be checked every time they are used, by aspirating and confirming a low pH, and only x-raying if this is not the case.
Seek specialist advice if the patient has a high risk of aspiration or any deviation to normal anatomy
Organisations and managers can make systems safer by:
- identifying a clinical lead to implement actions
- reviewing existing policies and training and competency frameworks (eg, ensure a doctor with suffi cient seniority is responsible for signing off the use of NG tubes)
- ensuring stock of correct equipment (approved pH indicator paper and radio-opaque tubes with clear length markings)
- restricting procedures done out-of-hours.
MPS strongly advocates mandatory documentation of the method by which the NG tube’s position is confirmed. Documenting confirmation of correct placement should safeguard against accidental and potentially catastrophic use of NG tubes.
Further clinical research is needed in this area, but small studies have suggested that magnet-tracking devices, where a magnet is inserted into the tube tip, may hold promise for the future.5 In the meantime, no existing bedside methods are completely reliable in testing the position of NTs, so being mindful of the complications will mitigate the dangers.
Thanks to Sir Liam Donaldson and Dr Sukhmeet Panesar from the NPSA for their help with this feature.
- Blunden M, Hinton J, Star Wars and Harry Potter musician died after ‘doctor put food tube into his lung’, London Evening Standard (27 Jan 2011)
- NPSA, Reducing Harm by the Placement of Nasogastric Feeding Tubes (Feb 2005)
- NPSA, Harm from Flushing of Nasogastric Tubes Before Confirmation of Placement (March 2012)
- Mayor S, NHS extends never events list and introduces cost penalties, BMJ (2011;342:d1263)
- Bercik P, Schlageter V, Maruo M, Rawlinson J, Kucera P, Armstrong D, Non-invasive Verification of Nasogastric Tube Placement Using a Magnet-Tracking System, JPEN J Parenter Enteral Nutr (2005, 29(4):305-10)
Download a PDF of this edition